Professional Documents
Culture Documents
GALAXIES
Lauren V. Jones
GREENWOOD PRESS
An Imprint of ABC-CLIO, LLC
Copyright by Lauren V. Jones
All rights reserved. No part of this publication may be reproduced, stored
in a retrieval system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, except for the inclusion of
brief quotations in a review, without prior permission in writing from the publisher.
This book is also available on the World Wide Web as an eBook.
Visit www.abc-clio.com for details.
ABC-CLIO, LLC
Cremona Drive, P.O. Box
Santa Barbara, California -
This book is printed on acid-free paper
Manufactured in the United States of America
To my husband, Tarek,
and my children, Omar Zakaryah and Leilah Sorayah.
I love you more than the whole universe!
This page intentionally left blank
Contents
Series Foreword ix
Preface xi
Acknowledgments xiii
Introduction xv
1 What Is a Star? 1
2 Star Light, Star Bright 37
3 Putting Together the Puzzle of Stellar Evolution 55
4 Variable Stars and Multiple Star Systems 85
5 What Makes a Galaxy? 107
6 The Milky Way 129
7 Arms in Disk Galaxies 145
8 Galaxy Interactions 161
9 Active Galaxies 171
10 Galaxy Evolution 181
Glossary 191
Bibliography 199
Index 201
vii
This page intentionally left blank
Series Foreword
Not since the 1960s Apollo-era has the subject of astronomy so readily cap-
tured our interest and imagination. In just the past few decades, the a con-
stellation of space telescopes, including the Hubble Space Telescope, have
peered deep into the farthest reaches of the universe and discovered super-
massive black holes residing in the center of galaxies. In concert, giant tele-
scopes spread around the globe on Earth’s highest mountaintops have spied
planet-like objects larger than Pluto lurking at the very edges of our solar
system and carefully measured the expansion rate of our universe. Simul-
taneously, meteorites from Mars with bacteria-like fossil structures have
spurred repeated missions to Mars with the ultimate goal of sending humans
to the red planet. Given that astronomers have recently discovered hundreds
of planets beyond our solar system, we are given pause and a rationale to cap-
ture what we now understand about the cosmos in these Greenwood Guides
to the Universe volumes as we prepare ourselves to peer even deeper into the
universe’s secrets.
As a discipline, astronomy covers a range of topics stretching from the cen-
tral core of our own planet outward past the Sun and nearby stars to the most
distant galaxies of our universe. As such, this set of eight volumes systemati-
cally does the same, covering the main components of our solar system (The
Sun; Inner Planets; Outer Planets; and Asteroids, Comets, and Dwarf Planets)
and all the major structures and unifying themes of our evolving universe
(Stars and Galaxies; Cosmology and the Evolution of the Universe; and Astron-
omy and Culture). Each volume comprises a narrative discussion highlight-
ing the most important ideas of major celestial objects and how astronomers
have come to understand their nature and evolution. In addition to describ-
ing astronomers’ most current understandings of the topics covered, these
volumes also include perspectives on the historical and pre-modern under-
standings that have motivated us to pursue deeper knowledge.
The ideas presented in these Greenwood Guides to the Universe vol-
umes have been meticulously researched and carefully written by experts to
provide students and interested non-expert readers with the most current
and scientifically accurate information and understandings of astronomers
ix
x • SERIES FOREWORD
xi
xii • PREFACE
The author would like to thank her family for their support throughout the
long process of writing this volume. Also, the author would like to thank
those who taught her the most about astronomy and how science works (in
chronological order): Dr. Profeta, Mr. Sweeney, Debbie Elmegreen, Henry
Albers, Cindy Schwarz, Fred Chromey, Mort Tavel, Emilia Belserene, Alan
Harris, John Gaustad, Anatoly Vladimirovich Zasov, Nikolai Nikolaievich
Shakura, Mikhail Sazhin, Bill Keel, Henry Emile Kandrup, Richard Joseph
Elston, Elizabeth Lada, Gus Muench, Tim Spahr, Karl Haisch, Tim Slater,
Larry Marschall, Ed Prather, and Esther Hopkins. Each of these individuals
contributed significantly to developing my ability to write this volume.
Finally, to those who served as role models for the author, she would like
to express her gratitude for their leadership and exemplary scientific integ-
rity here: Vera Rubin, Saul Perlmutter, Robert Kennicutt, Robert Kirschner,
John Huchra, and Dorritt Hoffleit.
xiii
This page intentionally left blank
Introduction
The study of stars probably began before recorded human history. From
an anthropological point of view, humans probably used the stars and
their regular patterns to predict seasonal changes. Archeoastronomy is
the modern science of searching for evidence of this, and archeoastrono-
mers have found significant evidence that ancient cultures monitored the
motion of the stars, sun, and moon for various purposes. The study of
galaxies is still in its infancy. Astronomers were not even aware that galax-
ies existed until the early part of the 20th century. Nonetheless, much has
changed since that realization. Astronomers have learned a lot in a very
short period of time.
Modern astronomy is an entirely different endeavor than astronomy in
any other time period. The only thing modern and ancient astronomers have
in common is their subjects. Ancient astronomers, of course, did not have
the tools we use today to gather the kind of data we now have about stars and
galaxies and their physical natures. Ancient astronomers were responsible
for mapping the positions of the stars and noticing that some moved rela-
tive to others, distinguishing them from the other stars as “wanderers,” or
planets.
Early modern astronomers built on that knowledge and used telescopes to
find other kinds of objects. They noted all the “fuzzy” objects, or nebulae, that
they could find and recorded their positions. They began to be able to see the
surface and cloud features of the planets and could distinguish the shapes of
some nebulae as spiral. For the pursuit of the study of stars, the most impor-
tant discovery in early modern astronomy was that of clusters of stars.
In the early 20th century, the telescope was the main tool used by astrono-
mers; however, by the 1920s, the spectroscope was widely used. The spectro-
scope allowed modern astronomers to study more than just points of light
in the night sky. Now astronomers could study the light itself. This is how
astronomers were able to begin the discovery of the physical nature of stars
and the existence of the “island universes” we now call galaxies.
Because of the ability to collect data on photographic plates, astronomers
were able to collect more data than they could analyze. Suddenly, there were
xv
xvi • INTRODUCTION
more data than there were astronomers to analyze the data. This marked the
advent of women entering the field of astronomy. Many women were edu-
cated in the early 20th century, but there were few places an educated woman
could work. Astronomy was in need of educated workers, and there were
plenty of women who wanted to be involved in astronomy.
In the United States, Vassar Female College (now known as Vassar Col-
lege) had one of the most productive astronomy programs, led by the first
American female astronomer, Mariah Mitchell. Many women graduated
from her program at Vassar College and went on to work at the Harvard
Observatory as “computers.” Their contribution to the field of astronomy is
enormous and somewhat untold. Many have heard of these human comput-
ers, the most famous of whom is Annie Cannon, but few know of their many
accomplishments.
As a result of this frenzy of analyzing astronomical data from stars along
with the work of Ejnar Hertzsprung and Henry N. Russell, astronomers
pieced together stellar evolution. Many of the details of stellar evolution are
still unknown, but the big picture is fairly clear. More technological advances
have contributed to the detail we now possess.
The 1950s saw the advent of radio astronomy and the beginning of digi-
tized data. Soon after this came infrared astronomy, ultraviolet astronomy,
x-ray astronomy, charge coupled devices (CCDs, or digital cameras), and
much more. Digitized data meant that more data could be accumulated and
the storage of it would take far less space. Digitized data is also more objec-
tive, and the advent of easily attainable digital data revolutionized the study
of astronomy. No longer did astronomers rely on a person’s ability to “fly-
spackle” (compare the sizes of silver oxide spots on a photographic plate).
Digital data could be analyzed objectively in an automated way by modern
computers (not people, this time!).
Using almost the entire electromagnetic spectrum to study objects in the
sky, astronomers now know much more about the physical nature of stars
and galaxies. Using data taken over the last 100 years, astronomers have
more insight into some of the details of stellar evolution and are beginning
to understand the principle role of merger events in galaxy evolution. Using
telescopes all over the globe and in outer space, astronomers can see more
clearly (literally!) what a star or a galaxy is and how it behaves over short time
periods.
Modern astronomers who study stars ask questions like the following:
How does dust form on the surface of carbon stellar remnants? How does the
T Tauri phase affect the formation of planets and asteroids in a distant solar
system of planets? What is the maximum mass a star can have before it blows
itself apart? What is the minimum mass required for a remnant to become a
black hole instead of a neutron star?
Modern astronomers who study galaxies ask questions like the following:
What does a galaxy’s shape tell us about its physical nature? How much can a
galaxy change because of interactions with other galaxies? What is the nature
Introduction • xvii
of the central engine that powers active galactic nuclei? What can clusters of
galaxies tell us about galaxy evolution?
Despite almost a century of modern technology helping astronomers
study stars and galaxies, there are still many questions to be answered. The
study of stars and galaxies is an exciting part of astronomy, and it is, perhaps,
one of the longest-lived areas of modern astronomy.
This page intentionally left blank
1
What Is a Star?
Many have wondered, as these characters from the movie The Lion King
have, what the stars are. One way to try to determine what stars are is by
watching them move. Anyone can do this experiment. Just draw a picture of
what you see when you go out on a clear night. Be sure to include any objects
on your horizon as reference points. Then wait about an hour or more and go
back out to look at the sky. Check your drawing against what you see.
Because of the rotation of Earth on its axis, the stars appear to rise in the
east, move high in the sky (although not necessarily to a point directly over-
head), and then set in the west. This is true, no matter where you are in the
world. (Although, east and west are difficult to define at the north and south
poles.) In the northern hemisphere, stars rise at an angle to the eastern horizon.
So, stars appear to rise high in the southern sky (not to a point overhead).
1
2 • STARS AND GALAXIES
Figure 1.1 This is a diagram of a celestial sphere. A celestial sphere is a tool used in astronomy to rep-
resent the motions of objects in the night sky due to Earth’s rotation on its axis. In this model, Earth
is a stationary point in the center while the Sun and stars appear to move around it. This is consistent
with what we observe on the surface of Earth, but not representative of the actual motions of bodies.
From Earth, in reality, the Sun and star only appear to move because Earth is rotating on its axis.
[Jeff Dixon]
In the southern hemisphere, the stars rise in the east as they do everywhere
else on Earth, but before it sets in the west, a star will rise to a point high in
the northern sky (not to a point overhead). It is disorienting to someone who
has only seen the northern hemisphere sky to observe the sky in the southern
hemisphere (and vice versa, of course). Even to those who do not usually pay
very close attention to the sky, something seems “not right.”
The only place on Earth where a star can rise due east and pass through the
point in the sky directly overhead before setting due west is at the equator. At
this point on the surface of Earth, a star that rises due east will follow a path
through the sky will take it through the zenith (otherwise known as the point
in the sky directly overhead).
What Is a Star? • 3
The stars appear to move all together. One can divide the stars into groups
(called constellations) that maintain their shape and relative positions
throughout the night. This is because the stars are not actually moving. The
stars only appear to move because Earth is rotating on its axis. In fact, the
motions of most stars relative to the Sun are so small that it takes thousands
of years for almost any visible change to manifest itself. The motions of stars
relative to the Sun are so small because the stars are so far away.
Constellations
Constellations have been defined and described by many different cultures in the history of human-
kind. The constellations used by astronomers today were, for the most part, designated by the ancient
Greeks. There are 88 constellations defined by the International Astronomical Union. Those in the
northern hemisphere are Greek in origin, but in the southern hemisphere some constellations were
given their boundaries and names by the early explorers. Most of the Greek constellations have myths
associated with them. These stories tell of the creatures or characters they represent and how they
ended up in the night sky (for example, see Ovid’s Metamorphosis which describes the mythological
origin of most of the Greek constellations). While other cultures have grouped the stars into different
constellations, it is the Greek names that were adopted by modern astronomers. Interestingly, the
constellation Orion (the Hunter, in Greek mythology) is similar in many different cultures. It has
similar boundaries, as well as a similar form. Many cultures (including, Chinese, Egyptian, Native
American cultures) designated the stars in this region of the sky as representing a human man.
Some of the so-called constellations with which we are most familiar are not constellations at all.
For example the Big Dipper is not a constellation. It is an asterism, or group of stars that are part of
one or many constellations. In the case of the Big Dipper, the stars that make up the asterism are part
of the constellation Ursa Major. Another well-known asterism is the Summer Triangle. Stars that
make up the Summer Triangle are from three different constellations: Lyra, Cygnus, and Aquila.
Studying the relative motions of stars merely tells us that they are very far
away. Nearly all stars stay in the same place relative to one another night after
night, month after month, year after year. Very few stars appear to move rela-
tive to the others. Stars that move relative to others are called high proper
motion stars. These stars are very close to our Sun, compared to the oth-
ers, and are moving exceptionally fast—relative to our Sun’s motion through
space.
Several objects look like stars and also appear to move relative to the nearly
constant stellar background; these are the planets. The word “planet” comes
from the Greek word for “wanderer.” The planets were so named because
they appeared to move relative to the other stars. At that time, the Greeks
did not know that the planets and the stars were fundamentally different
objects; they were merely noting the different motions of the two objects.
The motions of the planets are so significant that one can observe changes in
positions of the three nearest planets (Mercury, Venus, and Mars) on at least
a weekly scale. That is, one might not notice changes from night to night,
but if one looks at a planet’s position relative to the stars this week and then
4 • STARS AND GALAXIES
looks again next week, one will probably be able to detect that the planet has
changed position with respect to the stars.
The positions of objects in the night sky are designated by coordinates
called right ascension and declination. Right ascension is like longitude
on Earth; it is a measure of how far a star is behind the constellation Aries,
in units of hours, minutes, and seconds of sidereal time. The constellation
Aries marks the “prime meridian” of the sky. On Earth the prime meridian
is the line that goes through Greenwich, England. In this location on Earth,
there was an observatory, and time on Earth was kept by this observatory
for centuries, so it became the reference line to the lines of longitude on the
globe. In the sky, the path of the Sun relative to the stars crosses the celestial
equator (a projection of Earth’s equator on the sky) in two places: the Vernal
Equinox and the Autumnal Equinox. Our calendar year used to begin on
the Vernal Equinox, so it was chosen to be the reference line on the celestial
sphere.
The time it takes for Earth to rotate once on its axis relative to the stars is
about 23 hours 56 minutes. This is less than what we call a “day” on Earth.
The reason that it takes less time for Earth to rotate on its axis relative to the
stars than it does to rotate on its axis relative to the Sun is because of its orbit
around the Sun. As Earth rotates on its axis it moves in a nearly circular
orbit around the Sun. This means that Earth must make a little more than
one full rotation to get the Sun to the same relative position in the sky. So,
the solar day is longer than the sidereal day by about 4 minutes.
Thousands of years ago, people who studied the stars and recorded their
positions were called astrologers. Nowadays, astrologers still track the posi-
tions of the planets relative to the stars. Then and now, astrologers believe
that the relative positions of stars and planets have some meaning in the lives
of humans on Earth. Although science has found no evidence for such a con-
nection, modern astronomers often use the data recorded by ancient astrolo-
gers to learn more about the objects astrologers watched.
One thing the ancient astrologers did exceptionally well was to carefully
note the times and positions of irregular phenomena. It is because of ancient
astrologers’ records that the periodic nature of comets was discovered. It is
also ancient astrologers’ records (in combination with modern astronomi-
cal studies) that tell us how the Crab Nebula came to be. Ancient astrol-
ogers were the first record keepers and the first data gatherers in the field
now known as astronomy. While modern astrologers do not contribute to
modern astronomy, there is an interesting overlap between the two practices
thousands of years ago, before science truly existed.
Modern astronomers study the stars, and other objects in space, to better
understand how and why they are what they are. Astrologers are not inter-
ested in what stars and planets are, but where they are in the sky; they believe
that this has some relevance to human lives. There is, however, no scientific
evidence that the relative positions of planets and stars have any influence on
human lives.
What Is a Star? • 5
So what is a star? Studying stars’ relative motions only told us that the
stars are very far away. Stars do not move significantly, for the most part,
relative to one another. A human, or even a generation of humans, will not
detect any relative motions of most stars without the aid of modern instru-
ments of astronomy. The motions of the stars relative to one another, as well
as their apparent motions relative to Earth, did not tell us much about what
stars are. Until the birth of modern astronomy, there was no way to answer
the question of what a star is beyond mere speculation.
The closest star to Earth is the Sun. The details of what we know about the
Sun can be found in The Sun, another volume in the Greenwood Guides to
the Universe series. To sum up, the Sun is made up mostly of hydrogen gas,
with a significant portion of the Sun made of helium gas. A small percentage
of the Sun contains other elements in gas form. (So, Pumbaa was right. Stars
are giant balls of gas.) How we know what the Sun and other stars are made
of will be discussed in much more detail later in this chapter.
Even though the Sun is made entirely of gas, it is very dense at its core. The
Sun is not a solid body, like a planet, so even if it weren’t too hot, a person
couldn’t stand on its surface, because it doesn’t have a solid surface like the
surface of Earth. However, the density of the Sun near the core is very high,
much higher than the density of the Earth, even at its core. That extremely
high density is necessary for the Sun to produce light.
The Sun produces energy through the process of thermonuclear fusion
of hydrogen into helium in its core. This means, essentially, that hydrogen
atoms are being smashed together producing energy, as well as heavier,
helium atoms. Thermonuclear fusion will be discussed in much more detail
in chapter 3. After the energy has completed its long journey from the center
of the Sun to its edge, then through space, this energy is received by us on
Earth in the form of electromagnetic radiation (or, light).
Our Sun is huge, compared to Earth. The Sun has a mass of about 2 x 1030 kg,
which is about 330,000 times more massive than the Earth. The Sun has a
diameter of almost 1.4 million km. This is about 110 times the diameter of
Earth. If you compare the average density of the Sun to the average density
of Earth, you would find that the Sun has an average density that is about 1/3
that of Earth’s. Although the core of the Sun has a much greater density than
the core of Earth, the vast majority of the Sun has a much lower density than
even Earth’s atmosphere.
So, much is already known about the Sun. We have a sense of its size and
scale, as well as what it is and how it works. Still, there are some lingering ques-
tions: Are most stars like the Sun? (Most planets are not like Earth.) What are
the ranges of size, mass, brightness, and surface temperature that stars can have?
Do all stars have the same mechanism for producing energy in their cores? Do
6 • STARS AND GALAXIES
Figure 1.2The diagram on the left shows the relative mass of the Sun to Earth. The Sun is about
300,000 times more massive than Earth. The diagram on the right shows the relative sizes of the Sun
and Earth. The Sun is so large that about 110 Earths could fit across its diameter. [Jeff Dixon]
stars ever “die” or cease to exist? If so, how are stars “born” or formed? These
and other questions will be answered in this chapter and in this volume.
It turns out that quite a lot of stars are like the Sun; however, there are many
stars that have different sizes, masses, brightnesses, or surface temperatures.
When astronomers talk about other stars, they always use the Sun as a refer-
ence. For example, the mass of a star is usually given in solar masses. A solar
mass is the mass of the Sun (1.989 x 1030 kg). The symbol for a solar mass is
M. The solar radius (R) is 6.96 x 108 m. The brightness of stars is usually
discussed in terms of luminosity, which is a measure of the total light emit-
ted by a star, rather than the light received from a star. The solar luminosity
(L) is 3.827 x 1033 erg/sec or 3.827 x 1026 Watts (that’s much brighter than the
typical lightbulb!). The surface temperature of the Sun is 5,500 K. (Here, the
K is for Kelvin, a measure of temperature that is similar to the Celsius scale.)
Surface temperature is one of the few characteristics of stars that are never
used in reference to our Sun.
What Is a Star? • 7
The theoretical lower limit for a star’s mass is 0.08 M. If the mass of
the star is lower than 0.08 M, its core should not be dense enough or hot
enough to fuse hydrogen. Since it will not be able to fuse hydrogen in its core,
it will not be able to produce energy or the force necessary to balance the star
and prevent gravitational collapse. This lower limit is a calculated limit based
on what astronomers know about stars and how they work. There exists an
object, currently classified as a main sequence star (fusing hydrogen in its
core) that has a mass of 0.04 M.
The theoretical upper limit for a star’s mass is as yet unknown. The current
accepted limit is anywhere from 100–120 M. Astronomers hypothesize that
there must be a physical upper limit since the existence of a star is entirely
a balancing act between two forces. At some point, it is predicted that the
outward forces (balancing gravity) should be too strong and cause the star to
blow apart. The highest mass star ever observed (HD 92350) has a luminosity
corresponding to a mass of about 120 M.
Star Names
Star names are given in one of three ways. Most stars have at least one name, corresponding to the
catalog in which it appears. If the star is bright enough to be seen with a naked eye, it may have two
8 • STARS AND GALAXIES
names (one from a catalog and one from the constellation in which it is located). Very bright stars
usually have proper names also.
The brightest stars in the northern hemisphere sky were named by the ancient Arabian, Greek,
or Roman astronomers. The vast majority of the names with which many of us are familiar are
Arabic (like Vega, Betelguese, Rigel, Altair, and Aldebaran). These names were given to the stars
when astronomers of Arabia adopted the Greek constellations, but gave their own names to the
stars within them. We know this because, generally, the Arabic name has a meaning that is relevant
to the Greek constellation. Some stars, however, have retained their Greek or Latin names (like
Polaris, Sirius, Capella, Castor, and Pollux). There are at least several hundred stars with proper
names like these.
In addition to proper names, all the stars one can see with the naked eye (and some that
require a telescope to view) have names that are determined by the constellation in which it
resides and the relative brightness of the star within its constellation. These names are known
as a star’s Bayer designation. For example, Sirius is the brightest star in the constellation Canis
Major, so the Bayer designation for this star is Alpha Canis Majoris. Since “alpha” is the first let-
ter in the Greek alphabet, the use of “alpha” in the Bayer designation indicates that the star in
question is the brightest star in its constellation. The constellation name is changed slightly. The
original name is Canis Major, but in the Bayer designation it is Canis Majoris. The change is to
the possessive case in Latin. This means that a literal translation of the words “Canis Majoris” is
“of Canis Major.” The Bayer designation for Sirius is, then, the brightest star in Canis Major, or
Alpha Canis Majoris.
If the star is not the brightest star in the constellation, the letter of the Greek alphabet will be
different. For example, Bellatrix is the third brightest star in the constellation Orion. So, its Bayer
designation would be Gamma Orionis. “Gamma” is the third letter in the Greek alphabet. For con-
stellations that contain more visible stars than the Greek alphabet will allow, modern astronomers
have extended the Bayer designation by using the English alphabet and a double letter designation in
English to include even fainter stars.
Finally, the vast majority of stars that are not easily seen with the naked eye have names that are
related to a catalog where they are listed. There are many catalogs of stars. The most well-known
catalog that lists faint stars is the Bonner Durchmusterung catalog. This catalog was compiled in the
19th century. Stars in this catalog have names that look like this BD+38o 3238. This is actually the BD
number for Vega. In the Bonner Durchmusterung, Vega is the 3,238th star in the declination range
between 38 and 39 degrees north (positive). Another commonly used catalog is the Henry Draper
Memorial Catalog. This catalog lists stars by right ascension, rather than declination, but assigns
each star a number. In the Henry Draper Catalog, Vega is HD 172167. There are many catalogs that
list stars, and some stars (like Vega) have as many as 40 names.
Surprisingly, the smallest sized stars are not the least massive stars. The
smallest known stars are compact remnants of very large stars. The small-
est stars are called neutron stars. These stars no longer fuse anything in
their cores and are no longer made of hydrogen and helium. Neutron stars
can be as small as 10 km across. That is about 140,000 times smaller than
our Sun.
The largest stars are also nothing like our Sun in terms of their composition
or their method of producing energy. These stars are known as supergiants.
The largest known supergiant star (R Cassiopeiae) has a radius 1,800 R.
What Is a Star? • 9
BRIGHTNESS OF STARS
The brightest known stars have luminosities up to 100,000 times the solar
luminosity. These stars are supergiant stars: they are much larger (and can be
much hotter) than our Sun in addition to being much more luminous. Also,
supergiant stars produce energy using thermonuclear processes, but not in
the same way the Sun does. Supergiant stars fuse hydrogen into helium in a
shell surrounding a core which fuses other, heavier elements (like helium)
into even heavier elements (like carbon, oxygen, and nitrogen). As the super-
giant ages, its core gets hotter and denser allowing even heavier elements to
fuse. This continues up to the fusion of chromium into iron. After the chro-
mium fuel in the star has been used to create an iron core, the process stops.
The reason for this is that the process of fusing iron or fissuring iron (break-
ing it apart) takes energy, rather than producing energy. Since no energy
is produced by changing iron (either into to a smaller nucleus or a larger
nucleus), the star becomes unbalanced again. No energy is being produced,
so the force needed to balance gravity and prevent collapse is gone and a
catastrophic event, called a supernova occurs. Supernovae will be discussed
in much more detail in chapter 3.
The least luminous stars are about 100,000 times less luminous than the
Sun. It may be a surprise to learn that the least luminous stars are not the
smallest stars. Rather, these stars are very low mass stars that produce energy
essentially the same way our Sun does—through the process of thermonu-
clear fusion of hydrogen into helium. These stars have low luminosities pri-
marily because they have very low surface temperatures, but also because
they are small (but not the smallest). The relationship between luminosity
and size and surface temperature will be discussed in much more detail later
in this chapter.
TEMPERATURE OF STARS
The hottest known star (zeta Puppis) has a surface temperature of almost
40,000 K. That is just over seven times the Sun’s surface temperature. This
kind of star is producing energy by fusing hydrogen into helium through
thermonuclear fusion in its core, but the details of the process are a little dif-
ferent from the process in the Sun. This star is much more massive than our
Sun, and, therefore, has much more hydrogen to fuse; also, it uses up this vast
supply at a much faster rate than our Sun does.
The coolest known star (eta Muscae) has a surface temperature of just
under 3,400 K. This star is also producing energy by fusing hydrogen into
helium through thermonuclear fusion in its core, much like our Sun. This
cool star, however, is much less massive than our Sun and, therefore, has
much less hydrogen to fuse. Unlike the Sun, this star will fuse hydrogen with
a very high efficiency resulting in a longer life than one might expect for such
10 • STARS AND GALAXIES
a small star. For the star to live so long, it will fuse almost 100 percent of the
hydrogen it contains into helium, using convection processes to transfer fuel
from near the surface of the star into the core for fusion. Interestingly, the
range of surface temperatures of stars is much smaller than the ranges of the
other properties discussed.
COLORS OF STARS
Stars have been generally grouped into categories that describe their sizes.
There are giant stars and supergiant stars (all stars with very large sizes).
Generally, these are more luminous than the other stars, but their masses
and surface temperatures can fall almost anywhere in the ranges described.
The stars that are like our Sun, in terms of how they produce energy, are
called main sequence stars. These stars tend to stay close to one another in
size (usually not more than 10 times different from the Sun), but vary in sur-
face temperature and luminosity widely. Finally, the white dwarf stars (very
small in size) tend to be less luminous and at the same time, strangely hot
(very high surface temperatures).
What Is a Star? • 11
There are stars that are red, stars that are blue, stars that are yellow, and stars
that are white. How can stars have all these different colors? Why aren’t all
stars just yellow, like our Sun? Or, why aren’t all stars just white? Does a star
have the same color all the time? Are there stars that change colors?
Before we can really discuss why stars have different colors, we need to discuss
light. Light is the same thing as electromagnetic radiation. Electromagnetic
radiation is a phenomenon of nature that is caused by accelerating charged
particles. This phenomenon is called electromagnetic radiation because a
charged particle has an electric field associated with it (this is why a charge
repels like charges and attracts opposite charges). When a charge is moving,
the electric field creates a spontaneous magnetic field (this is why your com-
pass moves when you put it near a current-carrying electric wire and why a
gadget with magnetic memory, like a VHS tape or a memory stick, will be
erased if left near a strong electrical or magnetic source, like a television set or
a computer monitor). Because light is a result of an accelerating charged par-
ticle, and because accelerating charged particles have both electric and mag-
netic fields associated with them, light is called electromagnetic radiation.
Electromagnetic radiation, however, is not just the light we see with our
eyes. The light we see with our eyes is only a small fraction of what electro-
magnetic radiation actually is. When astronomers talk about the full range
of possible electromagnetic radiation, they call it the “electromagnetic spec-
trum.” Different kinds of electromagnetic radiation are categorized by their
wavelengths and/or energies. The wavelengths associated with visible light
(the light we see with our eyes) range from 400 to 700 nm.
There are many other kinds of electromagnetic radiation. Some, you may
have heard of before. Infrared is electromagnetic radiation with longer wave-
lengths and less energy than visible light (usually, this is considered to be the
light with wavelengths from 700 nm to 1,000,000 nm or 1 mm). Infrared radi-
ation is most often associated with the military use of night-vision goggles.
Night-Vision Goggles
The human body is at a temperature of about 300 K. An object with a temperature of 300 K emits
electromagnetic radiation that has a peak wavelength in the infrared part of the electromagnetic
spectrum. So, the reason night goggles work to see humans is because humans radiate electromag-
netic radiation and the peak wavelength of that radiation is in the infrared. That is, most of the elec-
tromagnetic radiation that humans radiate is in the infrared part of the electromagnetic spectrum.
12 • STARS AND GALAXIES
Black Lights
Ultraviolet radiation is also the source of so-called black lights. Black lights have been popular at
different times for use in personal homes and nightclubs. When black lights are used, white clothing
glows. This is because the white pigment in the clothing reflects all light. Black lights don’t appear
to give off much light because the peak wavelength of the light emitted is in the ultraviolet part of
the electromagnetic spectrum. Since the human eye cannot detect ultraviolet radiation, the human
eye does not perceive the majority of the radiation coming from a black light. The reason one can
see any light coming from a black light bulb is because the source of the light is an object that emits
light at all wavelengths, even though the majority of the light emitted is in the ultraviolet part of the
electromagnetic spectrum.
Photographic Film
Photographic film is a transparent film covered with a thin layer of photosensitive chemicals. Photo-
sensitive chemicals change their properties when light is incident upon them. In the case of photo-
graphic film, the photosensitive chemicals go from being transparent to being opaque. So, unexposed
photographic film is transparent. Once light is incident upon it, the film is chemically activated and
becomes opaque. This is the reason that negatives are the inverse of what is seen through the lens:
dark is light and light is dark.
The radiologist puts a piece of film behind the targeted area of your body. The
film detects the x-rays that pass through your body. Just like the film used in
a regular camera, the film is transparent until it is activated by some light. So,
when you see an x-ray, it is a negative image—dark where there was light and
light where there was darkness. An x-ray image shows the most dense parts
of your body.
On the long wavelength side of the visible part of the electromagnetic spec-
trum beyond infrared is microwave radiation. The wavelength range for the
microwave regime is from 1 mm to 1 m. This is the kind of radiation your cell
phone uses, as well as the kind of radiation used in microwave ovens.
The last and longest wavelength range is the radio part of the electromag-
netic spectrum. The wavelength range for radio is from about 1 m to 100,000
km. This is the wavelength range used to transmit television signals, as well
as the wavelength range used for the radio in your car.
Radio Waves
Radio waves are best for transmitting information over long distances. Their low energies make them
easy to modulate (so they can carry information) and cause them to reflect off clouds. This property
means that a radio signal can go a long distance and use the clouds to reflect the electromagnetic
radiation and keep it going back towards the surface of Earth.
So a good model for light might be a packet of energy that moves through
space (at the speed of light) like a wave. This is not an easy picture to visual-
ize, but then, light is not an easy phenomenon to describe or even define. It
is important to note, however, that light can behave both like a particle and
like a wave, although light itself is neither a particle nor a wave. A well-known
educator once compared the wave/particle duality to the platypus. The platy-
pus looks both like a duck and like a beaver, yet it is neither a duck nor a
beaver. Similarly, light behaves both like a particle and like a wave, yet it is
neither a particle nor a wave.
Since electromagnetic radiation behaves both as a particle and a wave, we
can talk about both the particle and wave qualities of electromagnetic radia-
tion. For example, we can talk about either the wavelength of electromag-
netic radiation or the energy of a photon. A photon is a particle of light.
In the discussion of different kinds of electromagnetic radiation above, the
definitions are based on wavelength. Electromagnetic radiation (or light) is
often classified by a wavelength, but it can also be classified by its energy. The
mathematical relationship between wavelength and energy is:
E=hc
λ
where h is the Planck constant, c is the speed of light in m/s, E is energy in
Joules, and λ is wavelength in m.
The most important part of this relationship is that energy and wavelength
are related inversely. That is, for larger values of λ (wavelength), E (energy)
gets smaller; and for smaller values of λ (wavelength), E (energy) gets bigger.
In other words, long wavelength photons have less energy than short wave-
length photons. If we put that in the context of the visible range of the electro-
magnetic spectrum, it means that red light has less energy than blue light.
Figure 1.4The above diagram shows two different blackbodies with different tem-
peratures. The hotter blackbody has a curve that peaks at a shorter wavelength than
the cooler blackbody. Since wavelength increases from left to right in this diagram,
the curve that peaks to the left is the one that belongs to the hotter blackbody and the
curve that peaks to the right is the one that belongs to the cooler blackbody. In this dia-
gram the hotter blackbody has a temperature of 25,000 K and the cooler blackbody has
a temperature of 5,000 K. [Jeff Dixon]
WIEN’S LAW
−6
λ = 2.9 × 10
T
Figure 1.5 In this diagram, blackbody B (right-most peak wavelength) has a lower tem-
perature and the same luminosity as blackbody A (left-most peak wavelength). Since
this can only be true if the cooler object is larger, the blackbody curves cross one another.
In cases where the curves do not cross, the blackbody sizes may be the same or very close
to one another. [Jeff Dixon]
total electromagnetic radiation. Does this sound impossible? Not really. The
amount of total electromagnetic radiation emitted by a blackbody is not only
dependent on its temperature, but also on its size. The next section of this
chapter discusses this relationship in much more detail.
L = 4πR2 σ T4
Here, L is the luminosity in J/s, the quantity 4πR2 is equal to the surface area
of the blackbody (presumably a sphere) in m2, σ is a constant (known as
the Stefan-Boltzmann constant) and T is the temperature of the blackbody
in K. There are no inverse relationships here, so all quantities are directly
related. This means that if we increase the luminosity of a blackbody, one of
three things could happen: first, the size of the blackbody could increase
and the temperature stays the same; second the temperature could increase
while the size stays the same; and third, both the size and temperature could
change in a way that causes the luminosity to increase.
It is not true that both the size and temperature must increase if the lumi-
nosity increases. The exponents in the relationship are an indication that this
relationship is more complicated than the others presented in this chapter.
Since the luminosity is proportional to the radius to the second power and
also proportional to the temperature raised to the fourth power, we have to
take these powers into consideration in our analysis.
Raising a number to the fourth power is the same as raising to the second
power two times (squaring a square). So a small change in temperature will
What Is a Star? • 19
result in a much larger change in luminosity than the same change in size.
That is, the luminosity is much more sensitive to changes in temperature than
changes in size.
To make this a little more concrete, let’s use some numbers. Say, blackbody
A has a temperature twice that of blackbody B, but they have the same size.
How will their luminosities compare? Well, to figure this out, we will look at
the Stefan-Boltzmann Law for each blackbody:
LA = 4πRA2 σ TA4
LB = 4πRB2 σ TB4
LA = 4πRB2 σ (2TB)4
LB = 4πRB2 σ TB4
Now we can relate LA and LB to one another. Since 24 is 16, we can factor that
out and we have
LA = 16(4πRB2 σ TB4) = 16 LB
So if we have two blackbodies of the same size and blackbody A has a tem-
perature twice that of blackbody B, blackbody A will be 16 times more lumi-
nous than blackbody B. What if blackbody A and blackbody B have the same
temperature, but blackbody A is twice as big as blackbody B?
Now we have
LA = 4π(2RB)2 σ TB4
LB = 4πRB2 σ TB4
Since 22 is 4, we can factor out the four and rewrite these equations so that:
So, for two blackbodies with the same temperature, the blackbody that is
two times larger will have a luminosity that is four times greater. Recall that
for two blackbodies of the same size the blackbody with a temperature twice
that of the other will have a luminosity that is 16 times greater. Therefore
for blackbodies, while luminosity depends on both size and temperature,
a change in temperature has a much greater effect than an equal change
in size.
20 • STARS AND GALAXIES
Figure 1.6
In this diagram there are four hotplates: two large and two small. One of the large
hotplates is set on a high temperature (shown as black) and one is set on a low temperature
(shown as light gray). The same is true for the smaller-sized hotplates. [Jeff Dixon]
STARS AS BLACKBODIES
visible part of the electromagnetic spectrum are red, orange, yellow, green,
blue, indigo, and violet. Technically, a star could be any of these colors or
any color in between. However, our eyes are not able to detect all of these
colors in light from distant stars. And, even though the peak wavelength of
a star may be green, if the peak is essentially flat within the visible part of
the electromagnetic spectrum, we will see white because all colors will be
equally bright. In fact, most of the stars we see appear to be white because,
even though the peak wavelength of the electromagnetic radiation coming
from the star is somewhere within the visible part of the electromagnetic
spectrum, the peak is essentially flat within the visible part of the electro-
magnetic spectrum.
On the other hand, when a blackbody curve is distinctly lopsided within
the electromagnetic spectrum, there is a clear differentiation of color. A star
will appear red or blue. (Actually most stars that are called “red” look quite
orange.) So, while in theory a star can have any color, in practice there are
really only four distinct colors observed in stars: red, blue, yellow, and white.
SPECTRAL CLASSIFICATION
Color was the easiest way to group stars, at first. Late in the 19th century,
however, telescopes were equipped with spectrometers. These instruments
allowed astronomers to separate the light coming from an astronomical object
into its many different wavelengths. Immediately, astronomers knew there
was more to learn about stars. In the late 19th century, it was thought that
stars were perfect blackbodies. The spectrum of a perfect blackbody should
look like a solid rainbow; however, when astronomers looked at the spectra
of stars, there were dark lines in the spectra. The spectra they observed were
not at all what one would expect for a perfect blackbody; however, the spec-
tra of other stars had some similarities to the spectrum of the Sun.
Absorption spectra caused other scientists to attempt to explain how the
dark lines could be formed, given what was understood about stars and the
Sun and how such spectra are formed. A culmination of many years of work
by many different and unexpected people (not all professional astronomers!)
unraveled the mystery of the dark lines in the spectra of stars and revealed
many new things about what stars are and how they work.
SPECTRA
To start out this discussion and to tell this amazing story, it is essential that
the notion of an astronomical spectrum is well understood. A spectrum (the
plural of spectrum is spectra) is an image of a light source taken through a
prism or some other light-dispersing medium. This kind of image is also
called a “spectrogram,” in astronomy. A spectrum shows the light separated
into its different wavelengths, like a rainbow.
The first scientist to attempt to understand the dark lines in a star’s spec-
trum was Joseph von Fraunhofer in 1814. Fraunhofer observed the dark lines
in the Sun’s spectrum through a prism. He made careful measurements of
the wavelengths of these lines and even designated them with letters. Fraun-
hofer did not understand why the lines were present in the Sun’s spectrum;
however, he determined that these lines were not an artifact of his prism, but,
rather, a natural part of the Sun’s spectrum.
another glazier, Philipp Weichelsberger. Through an unusual circumstance, only a few years later,
Fraunhofer met and was taken under the wing of the Prince Elector of Bavaria. The prince required
that Weichelsberger allow Fraunhofer time to study. Within a short period of time following this,
Fraunhofer went to work at the Optical Institute in Benectkbeuern (a monastery devoted to glass-
making). His study of the Sun began in 1814 with the invention of his own spectroscope. In 1818,
after having discovered how to make the best optical glass in the world and having invented very
precise methods for measuring the dispersion of light, Fraunhofer became the director of the Opti-
cal Institute. In 1821, Fraunhofer invented the diffraction grating, which allowed him to control the
dispersion of light and measure very accurately the wavelengths of the dark lines he observed in the
Sun’s spectrum.
In the 19th century, Robert Bunson and Gustav Kirchoff invented the spec-
troscope. This instrument was used to observe light from various sources.
With this instrument, scientists learned about emission and continuous
spectra. Bunson and Kirchoff discovered that every chemical element had
an associated set of dark or bright lines that identified the element uniquely.
They deduced, correctly, that the absorption lines seen in the Sun’s spectrum
were caused by the presence of certain elements in the atmosphere of the Sun
and the atmosphere of Earth.
In the early 20th century, the spectroscope was adapted to disperse the star
light received by telescopes, adding a photographic plate to record the spectra
observed. These instruments are called spectrometers. It is with the astro-
nomical tools of the 20th century that the spectra of hundreds of thousands of
stars were observed and studied, bringing to astronomy a deeper understand-
ing of the nature of stars.
Spectrometers
Refraction
Refraction is a property of waves. Waves are bent when they pass through a medium in which their
speed changes. So, light, behaving as a wave, travels slower in glass, for example, than in air. There-
fore, light passing through water is refracted. This is why a solid stick looks broken when viewed
above and below the water simultaneously.
bent different amounts. This is because what changes is not exactly the speed
of the photon, but the phase velocity of the wave front, causing the light to be
spread out by wavelength, like a rainbow. This effect is what makes a spectro-
graph “work”—this is what causes the spectrum.
The diffraction grating accomplishes the same end, but by a different
means. Diffraction is the bending of waves through interference of the
waves with one another. Interference causes some waves to cancel out with
one another while others add up. This creates a pattern of light and dark
regions for each wavelength. The bright regions for consecutive wavelengths
26 • STARS AND GALAXIES
Figure 1.8 The diagram above shows the parts of a spectrometer. The slit acts as an aper-
ture for the light, limiting the source to the light that can enter the slit directly. The dif-
fraction grating or prism spreads out the light and separates the light by wavelength. The
collimator lines up the rays of light so they are parallel. Finally, the detector measures the
number of photons at each point in the spectrum. [Jeff Dixon]
appear adjacent to one another making an image like a rainbow. So, in other
words, the diffraction grating separates the light by its wavelengths using the
interaction of light with itself. This effect is what causes the spectrum and
what makes the spectrometer work.
A diffraction grating is an optical element that allows light to pass through many (usually several
thousand or more) tiny (smaller than the wavelength of the light being transmitted) gaps. When light
waves pass through a diff raction grating the waves are bent and spread. In addition the light waves
pass-ing through a diff raction grating interfere with one another. Th e combination of all these pro-
cesses occurring is called diff raction.
Figure 1.9The above diagram shows each of the three different kinds of spectra. A con-
tinuous spectrum looks just like a rainbow of uninterrupted colors. This type of spec-
trum is formed by a blackbody source, like a hot solid or a hot, dense gas. An absorption
spectrum is a continuous spectrum, but with dark lines (absorption lines) where a small
range of specific wavelengths is missing. This type of spectrum is formed when the light
from a blackbody passes through a low density, lower temperature gas. An emission
spectrum is the opposite of an absorption spectrum; most of the length of the spectrum
is dark (no light), only lines of different colors are visible, and the number and colors
of the lines depend on the chemical composition of the emitting source. This type of
spectrum is formed in a hot, low-density gas. A blackbody may be the source of the heat,
but the gas must be hot enough that the atoms that make it up are mostly ionized or very
excited. In most cases in astronomy, this type of spectrum is indicative of matter in its
plasma state (ionized gas). [Jeff Dixon]
28 • STARS AND GALAXIES
Figure 1.10The above diagram shows the Bohr model of the hydrogen atom. This model is
named for its inventor, Neils Bohr. Bohr hypothesized that the electron of the hydrogen
atom had specific “allowed” orbits where it could be. Bohr’s model shows these “orbitals.”
The modern use of this less-than-accurate model is in the orbitals (now called energy
levels). Electrons can only have specific energies around their nuclei (for all atoms, not
just hydrogen). Bohr’s model allows one to calculate the wavelength of light associ-
ated with any energy level difference. Bohr’s model is also good for visualizing what is
meant by energy levels, although it is not an accurate physical representation of an atom.
[Jeff Dixon]
combine with the ionized atoms in the clou. When the electrons recombine
with the atoms, they lose energy in the form of light. The light emitted by the
electrons as they lose energy and step down the levels in the atom with which
they have combined, is directly related to the differences in energy between
any two allowed energy levels in the atoms of the cloud. So, if the cloud is
made of hydrogen, for example, the lines observed in the emission spectrum
will be at wavelengths that correspond to the energy differences between the
30 • STARS AND GALAXIES
Figure 1.11 The diagram above shows how each of the three different kinds of spectra are formed. A
continuous spectrum is formed by the emission of light from a blackbody (like a filament light bulb)
passing through a prism or grating. An absorption spectrum occurs when light from a blackbody
passes through an absorbing medium (like a cloud of gas whose atoms are not excited above the
ground state) and then passes through a prism or grating. An emission spectrum occurs when an
emitting medium (like a clouds of gas whose atoms are ionized or excited to a very high energy state)
emits light, which is then passed through a prism or grating. [Jeff Dixon]
Neils Bohr
Neils Bohr was born to Christian and Ellen Bohr of Copenhagen, Denmark, in 1885. Christian
Bohr was a professor of physiology at the University of Copenhagen, so Neils’s early interest in
science was nurtured and encouraged. At the age of 26, Neils Bohr received his doctorate. Fol-
lowing that, Bohr went to England to work with J. J. Thomson, but ended up in the lab of Ernest
Rutherford. Among the many things that Bohr is known for, one lesser-known accomplishment is
that he was the first to recognize was the importance of the atomic number. Bohr’s atomic model
stemmed from his idea that an atom could exist only in a set of discrete stable energy states. Bohr’s
model was able to predict the frequencies of the series of emission lines observed in the spectrum
of ionized atomic hydrogen. This model was the first step towards a quantum theory of the atom. In
Bohr’s atomic model, the frequency of the radiation emitted is proportional to the energy difference
between the two energy levels through which the electron has moved.
During World War II, Bohr and his immediate family were transported to Sweden, then to
England. There, Bohr worked with the British and American scientists who eventually developed the
atomic fission bomb. As part of the Atomic Energy Project, Bohr was intimately involved in creating
What Is a Star? • 31
the first bomb to exact massive human destruction. As a result, perhaps, he devoted much of his work
in his later years to peace and the peaceful applications of atomic physics. In an open letter to the
United Nations, he promoted openness and sharing of information as well as prevention of the use of
weapons of mass destruction.
allowed energy levels within the hydrogen atom. The number, brightness,
and colors of the lines in an emission spectrum depend on the atoms and
molecules out of which the cloud is made.
It follows from their descriptions that emission and absorption spectra
are the inverses of one another. That is, the lines missing in an absorption
spectrum are the lines that would appear if the same cloud were hot enough
to produce an emission spectrum. The lines in both the emission and absorp-
tion spectra tell us about the chemical composition of a cloud. The lines do
not tell us about the heating source (if it is an emission spectrum) or the con-
tinuous source (if it is an absorption spectrum). The lines only tell us about
the intervening cloud.
The spectrum of a star is, in fact, a combination of all three types of spectra
discussed above. However, for the most part, the emission lines of most stel-
lar spectra were not distinguishable in the first generation of stellar spectro-
grams. Instead, astronomers most often describe stellar spectra (even today)
as absorption spectra.
In the late 19th century, the director of the Harvard Observatory at the
time, Professor Pickering, realized and capitalized on the enormous potential
in the educated women then graduating from women’s colleges in the United
States. Most of the women who chose to pursue astronomy were unemployed
and looking for work. These women were well-trained and would work for
almost nothing, since they had no hope of ever being able to practice astron-
omy. One of the many women hired was Annie Jump Cannon.
Stellar Spectra
Stars have absorption spectra because their atmospheres contain elements that absorb the blackbody
source of light coming from the star’s “surface.” The atmosphere of a star is much less dense than
the star’s surface, so the atoms are less excited. Some atoms and molecules are even still neutral and
in low energy states, depending on the star’s temperature. What astronomers did not understand in
their analyses of stellar spectra early on, was that the elements they observed did not make up the
stars, but rather the stars’ atmospheres. And, what elements appeared in the spectra had more to do
with the surface temperature of the star than the composition of the star.
32 • STARS AND GALAXIES
Pickering assigned Cannon the tedious job of sorting stellar spectra. The
Harvard Observatory had been the recipient of the Henry Draper Memor-
ial fund. Pickering had long ago decided to put these funds to use in his
endeavor to obtain stellar spectra and classify stars by these spectra. Three
women had worked on the project previous to Cannon. Two of the three
developed their own classification scheme. Nettie Farrar started the project
in 1886, but worked on it for only six months before quitting to get married.
Figure 1.12
The figure above shows spectra of typical B, A, F, and G stars. In each spectrum shown,
the hydrogen absorption lines are labeled. These lines are strongest in the A stars, although they are
not the hottest. Other absorption lines begin to appear in the F and G type stars. This is because
the atmospheres of these stars are cooler, so heavier elements are present in lower energy states. At
higher temperatures, these elements are quicker to lose their outer electrons, which are the ones that
would cause visible wavelength absorption lines. [Jeff Dixon]
What Is a Star? • 33
Antonia Maury was the niece of Henry Draper. Upon graduating from Vas-
sar College in 1887, she was hired by Pickering at her father’s request. Maury
developed a classification system that was based on the strength and clarity
of the absorption lines found in stellar spectra. She also divided the stars
into groups based on which lines were visible. To this end, she developed
a method of line identification that drew mainly on what was then known
about the Sun’s spectrum, in addition to the spectra of several very bright
stars (in Orion, for example).
Maury’s classification scheme was complicated and relied on very good
spectrophotometry as well as a theoretical understanding for the cause of
the lines, which was not readily accepted or available at that time. (In fact, it
is not clear from the publication of the classification scheme that she under-
stood the underlying theory for why certain lines were present, while oth-
ers were not; however, her scheme was the first to separate main sequence
stars from giants and dwarfs based on the clarity and thickness of certain
absorption lines.)
Despite these facts, Maury published her classification scheme in a cata-
logue containing a fraction of the stars she classified, called “Spectra of Bright
Stars Photographed with the 11-inch Draper Telescope” in the Annals of the
Astronomical Observatory of Harvard College. Pickering wrote a forward
to the 134-page publication stating that Maury was responsible for the clas-
sifications in their entirety. Apparently, this was Pickering’s politically correct
way of stating publicly that he disagreed with Maury’s classification scheme.
Maury left the Harvard College Observatory near the time of the publication
of her work, which occurred in 1896.
When Cannon took over the job of classifying stellar spectra in 1911, she
developed a new classification system that was not unlike either Maury’s or
Fleming’s classification scheme. Cannon used Fleming’s letter groupings,
but regrouped them and reordered them based on what Maury had done.
Using Maury’s scheme, Cannon condensed Fleming’s 22 categories to about
10 distinct categories, and, in the process, Cannon also reordered Fleming’s
alphabetical categories. In the end, except for a few stars that turned out to
be spectroscopic binaries or other unusual phenomena, the spectral classifi-
cation scheme became O B A F G K M N R S. Further, Cannon developed a
way of identifying transitional spectra by adding a digit following each letter
(0–9) indicating which spectral type the star’s spectra was most like.
It is Cannon’s classification scheme that has survived and is used by mod-
ern astronomers. In her lifetime, Annie Jump Cannon classified over half a
million stars. To this day, she is the only person to ever achieve such a feat.
34 • STARS AND GALAXIES
Even though many have tried, there is not yet computer software that can
replicate the efficiency of Cannon and her team of human computers.
Although it was apparent that stars had different elements visible in their
spectra, it was still not clear to astronomers whether the stars’ spectra were
different because the stars contained different elements, or whether stars were
made of the same elements, but had different spectra for some other reason.
In the early 20th century, the accepted theory for planet and star formation
was that planets formed from material from the surface of the Sun. For this
reason, astronomers thought that stars must be composed of iron. At that
time, Earth was known to contain an iron and nickel core, so it seemed a
natural conclusion to draw that, given that the planets were formed from the
material on the surface of the Sun, the Sun must also contain iron (and other
heavy elements). But the spectrum of the Sun did not include evidence of all
the elements found on Earth, or even the same distribution of elements.
MEGHNAD SAHA
Meghnad Saha enters this story with the needed solution. He read the publi-
cations from the Harvard Observatory with a critical eye and tried to explain
what Maury and Cannon could not. Saha was born in 1893 in what was then
India, near the city of Dhaka (now the capital of Bangladesh). Saha was rec-
ognized early in his life for his genius. He went to a village elementary school,
then a city middle school, then the Dhaka Collegiate school. He ranked sec-
ond in the nation on the entrance exam for college and was admitted to the
Presidency College for studies in mathematics.
After accomplishing his masters of science degree in 1915 from Calcutta
College, Saha was asked to teach in the newly established Science College.
Although Saha wanted to study applied mathematics and physics, there was
no structure within the higher education system of his country to facilitate
that, so he continued his studies independently. His first major publication
in astrophysics was in the Astrophysical Journal in 1919. This publication
was his paper called “On Radiation Pressure and The Quantum Theory.” His
most famous scientific work “Thermal Ionization of Gases” was published in
1920. It is these two scientific papers that bring him into the story of stellar
spectra.
Saha found a solid theoretical basis for the spectral sequence developed
by Cannon. Saha’s 1919 dissertation on radiation pressure demonstrated
how the surface temperature of a star and the composition of a star’s atmo-
sphere would affect the absorption spectrum observed. A paper published in
1921, entitled “On a Physical Theory of Stellar Spectra,” explained this in the
What Is a Star? • 35
CECILIA PAYNE-GAPOSCHKIN
In the end, then, it was the work of four women at the Harvard Observatory
and one man from Bangladesh who solved the mystery of the dark lines
seen in the spectra of stars and the Sun. Three of the women were able to
see the patterns in the strange spectra and recognized a systematic change.
Documenting these changes was the first fundamental step toward under-
standing their meaning. The man from Bangladesh, along with the fourth
36 • STARS AND GALAXIES
woman, put the puzzle pieces together to reveal the underlying physics that
made all the work of the three women who classified the spectra come to
fruition.
Now, we understand that, although the spectra of stars have subtle differ-
ences, all stars are made of, basically, the same stuff: hydrogen and helium,
with traces of other elements. Also, we know that the subtle changes in
the spectra of the stars are due to differences in surface temperature, not
differences in chemical composition. We can actually determine the sur-
face temperature of a star, using the absorption line pattern found in its
spectrum. All this leads to even more information to process about stars. If
they’re all made of, essentially, the same things, why are they different tem-
peratures? Why do they have different colors? Why do they have different
luminosities? These, and other questions will be addressed in the following
chapters.
RECOMMENDED READINGS
WEB SITES
http://www.harvardsquarelibrary.org/unitarians/payne2.html
http://www.sdsc.edu/ScienceWomen/cannon.html
http://en.wikipedia.org/wiki/Antonia_Maury
http://banglapedia.search.com.bd/HT/S_0022.htm
http://www.columbia.edu/~ah297/un-esa/ws1999-letter-bohr.html
http://nobelprize.org/nobel_prizes/physics/laureates/1922/bohr-bio.html
http://en.wikipedia.org/wiki/Joseph_von_Fraunhofer
http://www.calcuttaweb.com/people/msaha.shtml
http://eo.nso.edu/MrSunspot/records
http://www.astro.wisc.edu/~dolan/constellations/extra/nearest.html
2
Star Light, Star Bright
As many people already know, the first star one sees (near the western hori-
zon as the Sun sets) is actually usually a planet (Venus or Mercury, in fact).
However, this nursery rhyme actually has something to do with the way the
astronomical magnitude scale (used to describe the brightness of stars) was
developed. The magnitude scale is used by professional and amateur astrono-
mers alike. Knowing a star’s magnitude tells you how long to expose a digital
imager to take an image of the star, or how dark the sky has to be to see the
star if you want to observe the star with the naked eye.
Since astronomy has been taught, students of astronomy have been per-
plexed about this unusual scale. For one thing, the scale is backwards, which
makes it counterintuitive. Secondly, the scale is logarithmic. Although most
of the human senses actually operate on a logarithmic scale, most humans
are not very comfortable with the mathematics of a logarithmic scale. The
first section of this chapter will be devoted to clarifying the complications
(mathematical and logical) of the astronomical stellar magnitude scale.
MAGNITUDES
Before the time when astronomers had the tools of modern astronomy that
allow them to measure brightness in units of energy per second, ancient
37
38 • STARS AND GALAXIES
Light Pollution
Light pollution is becoming a big problem for modern astronomers. When humans lit their homes
with candles or even oil lamps, the night sky was very dark. It was easy to see the Milky Way in the
northern hemisphere. Nowadays, one has to travel far from urban centers to view the Milky Way.
With the expansion of cities over the last several decades, getting far from urban light pollution is
becoming more and more difficult. Many people have never seen the Milky Way or the Big Dipper
(the most well-known asterism in the northern hemisphere). Most children who grow up with urban
light pollution are not even aware that there are more than a handful of stars to see in the night sky.
In an effort to light up the night, humans have taken away their ability to view the universe.
This has contributed (along with the invention of the internet and cellular phone technology) to the
feeling that Earth is a small, isolated world. It is only the recent explorations of our solar system by
robots, images from our space-based telescopes, and brief trips of astronauts to the International
Space Station that interrupt our feeling of isolation in this universe. This is, in no small part, due to
the fact that humans no longer see the universe in our night sky.
To show how serious the damage is, satellite pictures of the night side of Earth have been taken. In
these images, the major urban centers are apparent in every country. The United States of America is
defined by the eastern seaboard (which is lit from the Florida Keys to the tip of Maine), the northern
border with Canada (lit from Prince Edward Island to Vancouver), the west coast (lit from the north-
ern border of Washington to the southern border of California), and the southern border (lit from
San Diego, across the Mexican border, along the coasts of every southern state to the Florida Keys).
One cannot miss a single urban center within the United States. Every one is identifiable by a bright,
irregularly shaped blob on the map.
For professional astronomers this means that it is getting more and more difficult to find “dark
sites” on the planet. Observatories housing world-class telescopes built on mountains in the south-
western part of the United States are already suffering multiple effects of proximity to urban centers.
First, the pollutants in the air degrade the mirrors and their special coatings quickly, requiring them
to be recoated and repolished frequently. Second, the light pollution means that visibility at these
sites is reduced, requiring larger telescopes or more remote sites.
Star Light, Star Bright • 39
The first major observatory in the southwest was Mt. Wilson, located in the mountains just out-
side of Pasadena, California. This is where Edwin Hubble made many landmark observations that
changed the way astronomers understood the universe. This observatory was built in the low moun-
tains surrounding Los Angeles, west of Pasadena. Mt. Wilson is not used for astronomical research
any longer. It is surrounded by cell phone and radio towers and the light from the City of Angels has
made it impossible to view anything astronomically significant for many years. In fact, Hubble’s great-
est discovery at Mt. Wilson was made during World War II, when the city’s lights were turned off at
night to protect the city from attacks from enemies from across the Pacific Ocean.
Light pollution has had such a profound effect on astronomy that many astronomers have joined
an organization called the International Dark-Sky Association (IDA). This organization promotes the
use of lighting that does not interfere with viewing the night sky. Some countries have adapted their
recommendations in areas near astronomical telescopes to protect the science. For example, in the
Canary Islands, one can find some of the darkest skies in the modern world. The island of La Palma
enacted a law requiring that all lighting meet IDA standards. The telescopes housed on La Roque de
los Muchachos on the island of La Palma are becoming homes to some of the most important astro-
nomical instruments on Earth.
Apparent Magnitude
Two stars giving off the same amount of energy that are the same size, but
have different apparent magnitudes must be at different distances. Knowing
apparent magnitude and two out of the three qualities of energy output, dis-
tance, or size, will limit the remaining quality. For example, if two stars have the
same apparent magnitudes, give off the same amount of energy, and have the
same size, they must be at the same distance from Earth. And, two stars with
the same apparent magnitude, same size, and same distance from Earth must
be giving off the same amount of energy.
Absolute Magnitude
Absolute magnitude is a measure of how much energy a star gives off, or how
bright a star actually is. This is what is called “luminosity.” The total amount
of energy a star gives off is a measure of its luminosity. Luminosity is not
dependent on size or distance of the star. So, two stars with the same absolute
magnitude give off the same amount of energy, but they may be different
sizes or different distances from Earth. The absolute magnitude or luminos-
ity of a star does not provide any information about distance to the star or
radius of the star.
Alternatively, the absolute magnitude of a star is the same as the apparent
magnitude of the star if it were located at a distance of 10 parsecs (pc) from
Earth. The distance, 10 pc, was chosen as the standard because the logarithm
of 10 is 1. Mathematically, this makes the relationship between absolute mag-
nitude, apparent magnitude, and distance simpler. Also, 10 pc is a reasonable
distance for a star to be from Earth. There are more stars are at or beyond
10 pc from the Sun than there are stars closer than 10 pc from the Sun. (Inci-
dentally, since the distance between Earth and the Sun is so much smaller
than one parsec [it’s 1/206,265th of a parsec], astronomers assume that dis-
tance from Earth is the same as distance from the Sun for all stars and other
objects that reside outside the solar system.)
As noted before, the magnitude scale is backwards and logarithmic. The mag-
nitude scale is “backwards” because on the magnitude scale, a large num-
ber indicates a faint star. This is because the magnitude scale was devised by
watching as the stars appear after the Sun sets and the sky gets darker.
The scale is logarithmic because, interestingly enough, it turns out that
most human senses are logarithmic. We hear change in intensity logarith-
mically (hence the decibel scale), we see change in intensity logarithmi-
cally, we also detect odor and taste changes in intensity logarithmically.
So, although our brains don’t process the mathematics easily this way,
Star Light, Star Bright • 41
USING MAGNITUDES
What is commonly called “the inverse square law” is just another strange
phenomenon. This one is trickier than the magnitude scale, because at first, it
seems to make sense and agree with our day-to-day experience, but then the
inverse logic makes it difficult for our brains to process easily.
It starts out with something you already know: things that are farther away
appear fainter. This makes sense. We have all experienced this at one time
or another. It even fits with a basic rule that most humans learn early in
their existence: closer is more. It’s really the same thing: objects farther away
appear fainter means that objects closer appear brighter.
Now comes the hard part. The amount that the brightness changes by is
inversely proportional to the square of the distance between the observer
and the object. Sounds like a lot of math mumbo jumbo, right? Well, here it
is in a plain English example: Object A and object B have the same intrinsic
brightness, but object B is two times further from the observer than object A.
Object A will appear 22 = 4 times brighter than object B.
Let’s reason this one through step-by-step. Objects A and B have the same
intrinsic brightness. So, if they’re next to one another, one cannot tell which
is brighter, because they give off the same amount of light. But object B is
two times further from the observer than object A. Further means fainter, so
object B is fainter. But not two times fainter, two-squared times fainter, which
is four times fainter. Object B does not appear half as bright as object A, it
appears 1/4 as bright. Well, that’s a little unexpected. Didn’t we just learn that
our eyes detect light logarithmically? Why is it squared?
Imagine a point of light spreading out in all directions. At the source, the
light given off is the same in all directions. Move away from the light source by
one unit (meter, AU, light year, parsec, it doesn’t matter which). The light will be
fainter than on the surface. In every direction from the source, the amount of
light seen at that distance from the source (one unit) is the same. The light from
the source is spread out over that whole sphere. (Because we are dealing with
surface area, the number is squared. The area of anything is in square units.)
If we examine a square centimeter on the surface of a star and watch what
happens to the light contained in the original square centimeter as we move
Star Light, Star Bright • 43
away from the star, that area gets larger the farther we go out. (See figure 2.1
below.) With each unit of distance from the star, the area increases by a larger
amount. At one unit from the star, we have one area. At two units from the
star, the area increases to four times the size it was at one unit from the star.
At three units from the star, the area increases to nine times the size it was at
one unit from the star. If the light from one square centimeter is contained
in the first surface (at one unit from the star), then by the time we are three
units from the star, the light contained in that same area is now 1/9th the
brightness it was when it started.
It is the decrease in brightness over distance that is described by the inverse
square law. And, it is because of the inverse square law that astronomers use
absolute magnitude as well as apparent magnitude.
Recall that apparent magnitude tells how bright a star appears, while
absolute magnitude tells how luminous a star is, or how bright the star
would appear if it were at a distance of 10 pc. So, a star could have the same
apparent and absolute magnitudes. Such a scenario would occur if the star
were at a distance of 10 pc from Earth. If a star’s absolute magnitude is a
larger number than its apparent magnitude, that would mean that the star
appears brighter than it would be if it were at a distance of 10 pc. Con-
versely, if a star’s apparent magnitude is a larger number than its absolute
magnitude, that would mean that the star appears fainter than it would if it
were at a distance of 10 pc.
Figure 2.1 The above diagram depicts the relationship between distance and brightness.
As the light travels away from the source, the surface area increases as the square of
distance. So, if the light from a square travels to a distance two units from the source,
the radiation will be spread out over four squares, so that each square receives one-
quarter of the total light emitted by the same square at the surface. At three units from
the source, the light is divided into nine squares so that each receives one-ninth of the
total light emitted by a square on the surface. [Jeff Dixon]
44 • STARS AND GALAXIES
The two different types of magnitudes are related to one another. This is,
perhaps, apparent in their definitions. Mathematically, however, the relation-
ship looks like this:
m – M = 5 log d – 5
Parsec
The distance unit parsec is a word derived from two other words: “parallax” and “arcsecond.” The
parsec is defined as the distance to a star that has a parallax of 1 arcsecond. To understand, fully, what
this means, we must understand both the concept of parallax and the unit arcsecond.
An arcsecond is a division of a degree of arc (there are 360 degrees in a circle). A degree is divided into
60 arcminutes. Each arcminute is divided into 60 arcseconds. So, an arcsecond is 1/3,600th of a degree.
Parallax is a method for measuring distance to stars. This method is useful for measuring the
distance to only the nearest stars (out to a distance of about 100 pc, when measured from Earth). As
Earth orbits the Sun, nearby stars appear to change position relative to stars that are farther away. The
amount by which the nearby star’s position changes is related to its distance.
On Earth, humans have experience with the parallax effect. This is why, when you watch the
landscape as you sit in a moving car or train, the nearby landscape appears to move quickly, but the
distant landscape appears to move more slowly or not at all. One can experience parallax simply by
holding a pencil at arm’s length at the center of your vision and close one eye at a time. The pencil
will appear to move relative to the background in your line of sight.
To measure a star’s parallax, astronomers observe the star (at least) two times so that the observa-
tions are six months apart. Since the observations are six months apart, during the second observation
Earth is on the opposite side of the Sun from the first observation. This means that Earth is as far away
from the position it was in at the first observation as possible. This maximizes the baseline. The angu-
lar shift of the star’s position is its parallax shift. The star’s parallax is equal to half its parallax shift.
To see how this relates to the star’s distance, a little bit of geometry is needed. The distance from
the star to Earth is the hypotenuse of the right triangle, the distance from the Sun to Earth is the short
leg, and the distance from the Sun to the star is the long leg of the right triangle. The angle between
the long leg and the hypotenuse is the parallax. Using some elementary trigonometry, we can easily
derive the relationship that follows:
sin (parallax) = distance between Earth and Sun/distance between Earth and star
Since the parallax of even the nearest stars is in 10ths or 100ths of an arcsecond, the angle is very
small, so we can use the small angle approximation that says that the sine of a very small angle is
approximately equal to the size of the angle. Plugging in the distance between Earth and the Sun as 1
astronomical unit (AU), the equation now reads,
p = 1/d
where p is parallax (in arcseconds) and d is distance to the star. Since, by definition, a parsec is the
distance to a star with a parallax of 1, the distance is in units of parsecs.
Star Light, Star Bright • 45
In the early 20th century, astronomers were just beginning to learn about
the physical nature of the universe in which they lived. It took almost 5,000
years of written history for humans to understand that Earth was a planet in
a system of planets. At the turn of the 20th century, when Hertzsprung and
Russell were working, astronomers were just beginning to understand that
the universe was far more vast than anyone had ever imagined.
At the Harvard Observatory, the observatory director Professor Edward
Pickering had directed the women he employed to take spectra of every star
in the heavens and to classify those spectra by the absorption lines visible
within them (see also chapter 1). His goal was to understand what the spec-
tra of stars could reveal to astronomers about the physical nature of stars.
It turned out that he could correlate the information he gleaned from their
spectra to their surface temperatures and their sizes (sometimes). With the
help of other scientists who studied the physics of extremely hot and low-
density gases, astronomers were able to draw the conclusion that all stars
are made of the same ingredients: about 70 percent hydrogen, 30 percent
helium, and a smattering of other elements.
Hertzsprung wanted to know other things. He did not have all the spec-
tra that the women at the Harvard Observatory were studying, but he was
doing spectrophotometry (measuring brightnesses of stars in different parts
48 • STARS AND GALAXIES
of the visible spectrum) on his own and used star temperature (derived
from his spectrophotometry) and luminosity (in the form of absolute
magnitude) to create a diagram. Russell was studying stars and measuring
their distances from Earth to determine their absolute magnitudes. Russell
used these data along with the Harvard women’s spectra to make his own
diagram.
Both diagrams showed the same unusual feature. While most stars
appeared to be on a line that went from high temperature and high lumi-
nosity to low temperature and low luminosity, there seemed to be a second
area of the diagram (low temperature and high luminosity) where a signifi-
cant population of stars existed. Both men concluded that these two groups
were related and represented different stages in a star’s evolution. Both men
also concluded that since the stars with low temperature and high luminosity
were rare, this stage of evolution was a more rapid phase. And this is how the
H-R Diagram was born.
Scientists recognized that both Hertzsprung and Russell had come upon an
aspect of the physical nature of stars that would not be revealed by studying
the spectra of stars alone. The work of the women “computers” at Harvard,
in conjunction with the works of Annie Cannon, Megnad Saha, and Cecelia
Payne-Gaposchkin, were used to expand the H-R Diagram to learn more
about the physical nature of stars and to deduce stellar evolution. The details
of how these three individuals contributed to the modern understanding of
stellar evolution were detailed in chapter 1.
(from the blue end to the red end) as follows: violet, indigo, blue, green, yel-
low, orange, red. The human eye does not distinguish well between violet and
indigo, so most humans usually describe the end of the spectrum as “purple”
or “blue.”
A star’s temperature (on its surface) is related to its peak wavelength. This
is the wavelength at which most of the star’s energy is being radiated. Stars
that have their peak wavelengths in the blue end of the electromagnetic spec-
trum (the peak can be anywhere from the blue end of the visible part of the
spectrum to the ultraviolet part of the electromagnetic spectrum) appear
bluish. Stars that have their peak wavelengths in the middle of the visible part
of the electromagnetic spectrum usually appear white because these stars
emit most of their energy all throughout the visible part of the electromag-
netic spectrum. When all colors of light are mixed together, the light appears
white. Stars with peak wavelengths in the middle-red end of the visible part
of the electromagnetic spectrum appear yellow or golden. Finally, stars with
peak wavelengths in the red end of the electromagnetic spectrum (the peak
can be anywhere from the red end of the visible part of the spectrum to the
infrared part of the electromagnetic spectrum) appear red-orange.
On the H-R Diagram, along the x-axis, the temperature goes from hot-
test to coolest, with temperature increasing to the left. Knowing this and
the information given earlier, we can begin to piece together the physical
characteristics of a star by its temperature, color, or spectral type. An O star,
for example, has a temperature ranging from about 20,000 to 50,000 degrees
Kelvin, and will appear bluish-white. On the other hand, a red star must have
a spectral type K or M and a surface temperature of 2,000–3,000 K. Our Sun
has a surface temperature of about 5,500 K, it appears yellow-white and is a
spectral class G star. Knowing one of the three possible quantities that make
up the x-axis of the H-R Diagram tells one at least the range of the possible
values for the other two quantities. But wait, there’s more!
L*/L = 2.512(4.83-M*)
In this equation L* is the luminosity of the star in question, Lis the luminos-
ity of the Sun, and M* is the absolute magnitude of the star in question. The
above equation is valid for measurements using the V (visual) astronomical
filter.
Star Light, Star Bright • 51
The absolute magnitude and the luminosity of the star comprise the y-axis
of the H-R Diagram. Knowledge of either the absolute magnitude or the
luminosity of a star requires some knowledge of the star’s distance from Earth.
Apparent magnitude can be measured from a simple brightness measurement,
but distance is required to determine luminosity or absolute magnitude.
The fact that two pieces of information are needed to put a star on the
y-axis of the H-R Diagram means that three pieces of information are needed
to place a star on the diagram: temperature (or spectral class or color), appar-
ent magnitude (or brightness), and distance (to get absolute magnitude or
luminosity). From this information, it might seem that it would take a long
time to be able to put enough stars on the H-R Diagram to begin to see any
trends; however, because of the work so many astronomers were doing in the
early 20th century, there were already plenty of stars with sufficient data to
populate the diagram and learn much about the physical nature of stars.
Upon populating the H-R Diagram with stars, astronomers noticed some
very interesting features. The first, and perhaps most important feature
astronomers noticed, was that about 90 percent of the stars plotted formed
a line that goes from the upper left corner to the lower right corner of the
diagram. Since most stars fell on this line, it was called the “main sequence.”
The Sun, it turns out, is a main sequence star. Main sequence stars span the
full possible range in brightness, and temperature; however, using the Stefan-
Boltzmann relation, we can see that main sequence stars span a small range
of sizes from 10 times smaller than our Sun to 10 times larger than our Sun.
Because so many of the stars plotted fall on this line, astronomers sur-
mised that the main sequence line must be a very important piece of infor-
mation about how stars exist. However, it wasn’t until astronomers started
52 • STARS AND GALAXIES
to look at the H-R Diagrams of clusters of stars that they were able to piece
together the meaning of the main sequence. This discovery will be discussed
in greater detail in chapter 3. The main sequence of stars follows a straight-
forward mathematical relationship that describes the existences of the stars
upon it. Main sequence stars with high luminosities have high temperatures
and are larger and more massive than main sequence stars with low lumi-
nosities (which have low temperatures are smaller and less massive).
What astronomers now know is that the main sequence has temporal
(time) significance. It describes not just the majority of stars, but the major-
ity of a star’s existence. All stars begin their existence as stars on the main
sequence. Astronomers also now know that 90 percent of the existence of
a star will be on the main sequence. That is to say once a star leaves the main
sequence (once it changes temperature and size), 90 percent of that star’s
existence has been completed and the star will soon cease to exist.
Luckily for us, our Sun is still on the main sequence, indicating that it still
has a lot of time left before it changes temperature and size and begins its
journey to its end. Astronomers learned all this from the H-R Diagram when
they started to look at how clusters of stars appear on it. How astronomers
figured all this out and more detail about what we now know about stars will
be discussed in chapter 3.
THE GIANTS
In addition to the main sequence, three other main groupings of stars on the
H-R Diagram were noticed. One such grouping was named the giant stars.
Because these stars are located in the upper middle part of the H-R Diagram,
we know they are fairly luminous, and because they are on the right-hand
side of the H-R Diagram, we know they must have low surface temperatures.
So, their location on the H-R Diagram tells us these stars are luminous and
have a very low surface temperature.
Using the Stefan-Boltzmann equation, we can conclude that these stars
must be very large to give off so much energy with such low surface tem-
peratures. In fact, the giant stars range from 10 to 100 times larger than their
main sequence counterparts with the same luminosities. Since these stars are
on the right-hand side of the H-R Diagram, and therefore are red in color,
these stars are commonly known as red giants.
Again, as with the main sequence stars, this grouping of stars seemed
important (especially since there were no stars on the H-R Diagram outside
of these four main groups). The number of stars in the giant group was far
fewer than the number of stars in the main sequence, but stars appear in this
group, so it must be a significant phase in a star’s existence.
What astronomers know now is that the red giant phase is a phase that
many stars go through when they leave the main sequence. Since all stars that
have evolved past the main sequence have been through this phase, this is the
Star Light, Star Bright • 53
second most important group on the H-R Diagram. This phase does not last
long, however, so the stars in this group are not nearly as many as the stars
in the main sequence. Again, all this is information we know now because
astronomers spent so much time studying how clusters of stars appear on the
H-R Diagram.
THE SUPERGIANTS
A much smaller group of stars is located in the upper part of the H-R Dia-
gram spanning from the upper main sequence through the right side of the
diagram. Because these stars are in the upper part of the H-R Diagram, we
know that they are very luminous stars. Because they span from the upper
main sequence to the right side of the H-R Diagram, we know they have
lower surface temperatures than their main sequence counterparts. Again,
using the Stefan-Boltzmann Law, we can conclude that these stars must be
larger in size to produce such high luminosities with lower surface tempera-
tures. These stars are more luminous and can be much larger than the giants,
so they were named supergiants.
Supergiant stars are extremely rare, although there are a few in the neigh-
borhood of the Sun. These stars range from 10 to 1,000 times larger than
their main sequence counterparts with the same luminosities. These stars
distinguish themselves from the giant stars by their luminosities. The super-
giant stars are the most luminous stars, about 10 to 100 times more luminous
than the giant stars. Another significant difference between the giant and
supergiant groups is that there are supergiant stars with surface temperatures
much higher than the giant stars. This means there are both red and blue
supergiant stars, whereas there are only red giant stars.
Astronomers now know that supergiant stars were once the most luminous
and most massive main sequence stars. This phase of a massive star’s exis-
tence is very short, so very few stars exist in the supergiant group. However,
because these stars are so luminous, astronomers have probably observed all
that exist in the neighborhood of the Sun.
Astronomers cannot say this about populations of stars with lower lumi-
nosities. What makes a star easy to observe is its brightness. So, the most
complete samples of stars will include nearby stars with medium luminosi-
ties and all stars with high luminosities. That is to say, our understanding of
stars is limited by our ability to capture their light for study. So, the brightest
stars are the ones about which we have the most complete information.
The last of the main groupings of stars found on the H-R Diagram is a
group found in the lower left portion of the H-R Diagram. This group looks
54 • STARS AND GALAXIES
different from the previous two groups because it forms a narrow band par-
allel to the main sequence. Because this group is in the lower part of the H-R
Diagram, we know these stars are not very luminous. Because this group
is on the left part of the H-R Diagram, we know these stars have very high
surface temperatures.
If we use the Stefan-Boltzmann Law, we can easily see that these stars must
be very small to give off so little light at such high surface temperatures. This
group of stars was called the white dwarf stars. Because of their high surface
temperatures, these stars appear white. White dwarf stars are about 10 to 100
times smaller than their main sequence counterparts of the same luminosity.
Astronomers now know that white dwarfs are the final stage of existence for
stars that go through the giant phase.
RECOMMENDED READINGS
In the last chapter, intrinsic properties of stars and the H-R Diagram were
introduced. The H-R Diagram, in its inception, revealed the first clues to
understanding stellar evolution. Just from plotting the nearest stars and the
brightest stars, astronomers learned that stars fall into one of four general
categories. Astronomers identified the most common category as the main
sequence. And, early on, astronomers surmised that the second most com-
mon group (the giants) must be a stage in the evolution of stars, albeit a
shorter stage than the main sequence stage.
Currently we know much more about stellar evolution and even have cre-
ated models of what is happening in their interiors and how the light we see
is formed and transported from the core to the atmosphere of the star. How
did astronomers figure all this out? How can they be so sure their models are
correct? These are the questions this chapter will attempt to answer.
STAR CLUSTERS
The key to understanding stellar evolution came with the study of star clus-
ters. In the early 20th century, astronomers began to realize that some of the
nebulae they could observe were not just clouds, but stellar nurseries. When
astronomers realized that stars could form in groups, or clusters, they realized
the value in studying star clusters. Such clusters of stars would be approxi-
mately the same age and would be located in the same part of space (meaning
that they would all be the same distance away from the Sun).
55
56 • STARS AND GALAXIES
Even before the realization that some nebulae were stellar nurseries, clus-
ters of stars were well-known and catalogued. (What was yet to be known
about the nebulae was that stars formed this way—already in clusters, and
from the same materials.) Astronomers had identified two basic types of star
clusters: globular clusters and open clusters. Globular clusters are dense
clusters of stars in a nearly spherical formation. Open clusters are loose,
shapeless associations of stars.
Globular clusters contain about 10,000 to 1,000,000 stars. Usually little or
no gas or dust is found in a globular cluster. Globular clusters are gravita-
tionally bound systems of stars. These objects are relatively small, compared
to a galaxy. Where a galaxy, like the Milky Way, is about 30,000 pc across, a
globular cluster can be anywhere from about 3–60 pc across. (Compared to
our solar system, this is very large. Our solar system is only about 0.0004 pc
across!) So many stars in so little space means the stars are very close to one
another. Globular clusters are also mostly found above and below the plane
of the Milky Way. That is, if the Milky Way is shaped something like a pan-
cake, the globular clusters are like the syrup or the butter. They are very near
to the galaxy, but mostly they are not in the same part of the galaxy that our
Sun is in (the pancake part).
Open clusters (sometimes called galactic clusters because they are mostly
located in the disk of the Milky Way, and so appear in the galaxy) are loose
associations of stars. It is difficult to accurately define limits to the number
of stars or sizes of open clusters because even the most well-studied open
clusters may have members that have not yet been identified. Because of their
location within the plane of the Milky Way, open clusters are more difficult
to distinguish. Sometimes a few prominent members are obvious (like in
the case of the Pleiades), but the other members of the group may be more
difficult to identify. The plane of the galaxy is full of stars, and identifying a
cluster within the plane is difficult to do, unless you know each star’s distance
from Earth.
Unfortunately, for some stars, measuring distance is just not possible.
Some stars are too far away to measure their distances using parallax. While
there are other ways to measure distance, they require that the star be a spe-
cial type of variable star (discussed in much greater detail in chapter 4).
What is special about clusters is that all the stars within a cluster were
formed from the same cloud at approximately the same time, so that they
are all the same age. Also, because they are all in the same part of space, so
they are all the same distance from Earth. These small details were the key to
how astronomers were able to figure out stellar evolution.
Color-Magnitude Diagrams
Since the distance to most stars in clusters are not known, it is not possible
to put them on an H-R diagram; however, since the axes of the H-R diagram
are temperature and luminosity, astronomers can use other information
from stars in clusters to plot them on a similar diagram. Recall that color is
similar to temperature, so astronomers could, conceivably, use color, rather
than temperature or spectral type. In addition, recall that the absolute mag-
nitude of a star is its apparent magnitude, if it were at a distance of 10 pc.
Since all stars in a cluster can be considered to be at the same distance from
Earth, all the apparent magnitudes of the stars will be offset from the abso-
lute magnitudes by the same amount (proportional to their actual distance
from Earth, divided by 10 pc). So, astronomers plot a star’s color on the
x-axis (since color corresponds to temperature) and a star’s apparent mag-
nitude on the y-axis (since apparent magnitudes of stars that are all at the
same distance from Earth correspond to a star’s absolute magnitude). These
diagrams are called, therefore, color-magnitude diagrams.
58 • STARS AND GALAXIES
Figure 3.1Above is a color-magnitude diagram of a cluster of stars. The y-axis is the appar-
ent magnitude and the x-axis is color (the difference between the apparent magnitude
of the star in two different filters). The similarity of this diagram to the H-R Diagram
is what prompted astronomers to piece together stellar evolution. It was apparent that
clusters of stars were formed simultaneously and contained stars at various stages in
their evolution. The sequence of stellar evolution could be pieced together by studying
different clusters.. [Jeff Dixon]
EVOLUTIONARY TRACKS
Using the information gained by studying many globular clusters and open
clusters, astronomers were able to deduce the evolutionary sequence for stars.
Furthermore, from the data they gathered, they could trace evolutionary
tracks on the H-R Diagram showing how a star changes in physical appear-
ance over the length of its existence.
The first clue to stellar evolution is that the main sequence disappears
from the upper left. This means that the most massive, most luminous, hot-
test stars have the shortest existence on the main sequence. In fact, the data
show that the older a cluster is, the more diminished the main sequence is.
Astronomers used this information to create models that explained these
data. Knowing that main sequence stars produce energy through hydrogen
fusion, making some assumptions about dominant energy transfer processes
and the density of material at different points within the star, astronomers
can determine the length of time a star of a given mass will remain on the
main sequence. With this knowledge, the age of a cluster can be determined
by observing the location on the main sequence of the highest mass star (the
turn-off point) and then calculating the mass of that star and the length of
time it can fuse hydrogen. This length of time is the same as the age of the
cluster,
For example, take the well-known open cluster the Pleiades. This open
cluster is also known as the “Seven Sisters” and is found in the constellation
Taurus. While most people can see six or seven stars in this cluster, there are
actually more than 600 stars considered to be members of this cluster.
To determine both the distance to the object and the age of the cluster, a
color-magnitude diagram of the Pleiades is superimposed on an H-R Dia-
gram. The main sequence is then lined up on both diagrams so that the main
sequence of Pleiades aligns to the main sequence on the H-R Diagram.
Then, to determine the distance to the cluster, the vertical axis on the
color-magnitude diagram is matched to the vertical axis on the H-R Dia-
gram. Recall that the vertical axis on the H-R Diagram is absolute mag-
nitude. The vertical axis on the color-magnitude diagram is apparent
magnitude. So, when the two main sequences are matched up, the difference
between the absolute and apparent magnitude of any star on the diagram
Putting Together the Puzzle of Stellar Evolution • 61
can be determined. Hence, the distance can be calculated using the follow-
ing equation:
m – M = 5 log d – 5
At the tip of the red giant branch is where the transition from hydro-
gen shell fusion to core helium fusion happens for stars massive enough to
achieve this. The core of the star collapses a little, increasing its density, and
heats up to a much higher temperature and helium fusion begins. This new
energy source is in the core and not in a shell surrounding the core, so the
star shrinks back down to a smaller size (still larger than it was as a main
sequence star).
Stars that are stable and fusing helium in their cores enter the horizon-
tal branch increasing in temperature and decreasing in luminosity (moving
down and to the left on the H-R Diagram). As this occurs, some stars enter
the instability strip and become pulsating variable stars (discussed in great
detail in chapter 4).
If a star is massive enough to be able to fuse carbon after it runs out of
helium fuel in its core and shell, it will first follow the asymptotic giant
branch (which runs parallel to the red giant branch). The asymptotic giant
branch will take the star back to the giant group again, decreasing the surface
temperature and increasing in luminosity as it expands (moving up and to
the left on the H-R Diagram). Following this trip back to the giant group,
shell hydrogen and helium fusion begin, the star’s core collapses still more,
increasing its density and heating up to ignite fusion of carbon. Again, the
star stabilizes, igniting carbon fusion in the core, this time; the star shrinks
down (still remaining larger than the last time it shrunk) and heats up the
outer layer (moving downward and to the left on the H-R Diagram).
Each new fuel the star consumes allows the star to remain stable for a
shorter and shorter period of time. This is presumed by the small number
of stars found in this part of the H-R Diagrams of clusters. A small number
of stars in a region implies that the region is not occupied by stars for a long
time. (This is similar to the conclusion that because 90 percent of all stars
are on the main sequence, this is the state in which most stars exist most of
the time.)
From what is understood about energy production of fusion in terms of
atoms and atomic structure, astronomers know that stars can fuse elements
up to iron; fusing iron, however, requires energy to be added. Therefore, if
stars fuse iron, it will not help to keep the star in balance. It will cause the star
to collapse because without producing energy, there is no outward pressure
to balance gravitational collapse and the star will simply collapse.
From studying and trying to reproduce the fusion of elements in labora-
tories on Earth, astronomers know that the fusion of heavy elements (ele-
ments heavier than helium) requires extremely high densities of material
and extremely high temperatures. These conditions could, theoretically,
occur in the cores of stars, but do not likely occur in the outer parts of stars.
So, astronomers can calculate the amount of material that is needed so that
the star would be dense enough and hot enough to fuse heavy elements. The
amount of material available for fusion gets smaller and smaller with each
switch to a new fuel, which explains why each phase of the star’s existence
Putting Together the Puzzle of Stellar Evolution • 63
is shorter and shorter. It is estimated that stars that fuse silicon (which pro-
duces nickel, which rapidly decays into iron) do so for only about a day (this
time varies with the mass of the star, but not by much). Given that stars can
exist for millions and billions of years, fusing an element to sustain itself for
only one day is really a star’s final gasp.
So, stars evolve off the main sequence, and looking at clusters tells astrono-
mers much about how a star changes as it evolves. Studying very young star
clusters also reveals how stars become main sequence stars. In some color-
magnitude diagrams of very young clusters of stars, astronomers saw strange
groupings of objects that appeared very red. As astronomers began to observe
clusters using infrared detectors (which became more and more common in
the 1980s and 1990s), they noticed that in the youngest clusters, there were
a large number of these objects that could be plotted on color-magnitude
diagrams.
These strange objects turned out to be pre-main sequence stars (often
referred to as PMS stars). These objects are actually stars that are still in the
process of clearing away the debris from the clouds from which they were
formed. Some of these objects are also known as T Tauri stars.
T Tauri Stars
T Tauri stars are named after the first such object observed, an object in the constellation Taurus. By
the letter T designation, it is apparent that this object is one of the fainter stars in the constellation
Taurus. This star was discovered in 1852 by an astronomer named John Hind. At the time T Tauri was
noteworthy because it was an irregular variable star. A variable star is one that varies in brightness.
Astronomers were just beginning to realize that most stars that vary in brightness do so in a regular
and predictable fashion. This star, however, is one that is not regular. It turns out that the reason its
variation in brightness is not regular is because its brightness is varying for reasons that are very dif-
ferent from those stars which vary in brightness in a regular and predictable way. Regularly varying
stars vary their brightness, for the most part, by expanding and contracting. T Tauri stars are thought
to vary their brightness due to instabilities in their surrounding accretion disks.
T Tauri stars have spectral types G to M, so they appear to be early stages in evolution of low-mass
main sequence stars, like our Sun. These stars are thought to go through a phase in their evolution
where a disk of debris (from the nebula from which they formed) is surrounding the star. Some mate-
rial from the disk is accreted onto the star and may be spewed back into the outer parts of the disk.
Hotspots may occur in the disks of these stars and planets may form in the disks of these stars. Much
of the current research in stellar evolution and planet formation involves studying these kinds of stars
to better understand how they evolve and how our Sun and solar system may have come to be. Still,
much is not known, but models and observations are narrowing the range of possibilities for how
stars like our Sun may evolve.
64 • STARS AND GALAXIES
There are other populations of stars that are known to be PMS stars, but
the most interesting ones to astronomers right now, are the T Tauri stars.
This is probably because they offer so much insight into how our solar sys-
tem formed and how other solar systems may form. In astronomy, what is
interesting to astronomers is where the research will be. Therefore, very
little is known about higher-mass PMS stars. Recently, x-ray observations
may be revealing information about intermediate- to high-mass PMS stars.
It is still uncertain, however, as to whether these objects are high-mass PMS
stars.
From studying the youngest clusters of stars, astronomers have been able to
deduce pre-main sequence evolutionary tracks. These tracks are not entirely
supported by observations as the post-main sequence evolutionary tracks
are. Pre-main sequence evolutionary tracks are more highly dependent on a
small number of objects and extrapolation based on physics and what little is
known about the interior structure of stars.
Astronomers began to try to piece together the information they had into
a coherent picture of how a star comes to be. For stars less massive than
our Sun, the picture was fairly complete by observation alone, but for stars
like our Sun or more massive than our Sun, what happens before the objects
become main sequence stars is more mysterious. One could fairly easily use
mathematical models based on the low-mass stars and extrapolate to higher-
mass stars, however. And this is what has been done.
Hayashi pre-main sequence evolutionary tracks use both mathematical
and observational data to describe what happens to objects that are less than
one half the mass of our Sun before they become main sequence stars. Heyney
pre-main sequence evolutionary tracks use mostly mathematical and some
observational data to describe what happens to objects that are more than
one half the mass of our Sun before they become main sequence stars. Both
models imply a predictable sequence of events and a calculable period of
time for objects to become main sequence stars.
STAR LIFETIMES
As we learned above, stars exist for varying amounts of time. The amount of
time that a star exists depends on its mass. The amount of time that a star can
exist has everything to do with how and why stars exist in the first place. The
mechanism for producing energy inside a star is behind all this. If the star
produces energy, it exists. Otherwise, something will be out of balance and
the star will cease to exist.
Putting Together the Puzzle of Stellar Evolution • 65
In this section we will learn, in detail, how a star produces energy, how this
governs a star’s existence on the main sequence, and, for low-mass stars, how
this governs the rest of a star’s existence. The entire sequence of stages of a
star’s existence (for low-mass stars) will be outlined in detail here.
As you already know, this evolutionary sequence was determined by study-
ing clusters of stars of varying ages and deducing the evolutionary sequence for
stars of all masses. How this sequence is explained by the physics of the inte-
rior structure and mechanisms for energy production in stars will be revealed
in this section.
in the cores of main sequence stars involves beta decay. The nuclear reaction
can be written as follows:
p + p → D + e+ + νe
D + p → He3 + γ
where D is for deuterium, p is for proton, He3 is for helium-3, and γ is for the
gamma ray radiation.
To accomplish step three, steps one and two must have been completed
again so that there will be another helium-3 nucleus to interact with. In step
three, two helium-3 nuclei interact and form one stable helium-4 nucleus and
two protons. The nuclear reaction for this step can be written as follows:
where He3 is for helium-3, He4 is for helium-4, and p is for proton.
So, if we put together all three steps (including steps one and two, two
times each) the full reaction could be written as follows:
6p → He4 + 2p + 2γ
4p → He4 + 2γ
It is from this form of the equation that hydrogen fusion is usually taught.
Four protons come together and make a helium atom. Some energy is given
off because the mass of four hydrogen atoms is greater than the mass of one
helium atom. This is how astronomers can calculate the amount of energy
Putting Together the Puzzle of Stellar Evolution • 67
given off by this reaction. A hydrogen atom has an atomic mass of about
1.008 amu (atomic mass units = 1.66 x 10-27 kg), four times this would be
4.032 amu. A helium atom has a mass of 4.0026 amu. Four hydrogen atoms
have about 0.03 amu more than one helium atom. This very small amount
of “mass” is converted into energy using the famous equation, E = mc2. If
the proper conversions are done, the missing mass converted to energy is
equal to the energy of two gamma ray photons. So, even though the above
equation leaves out much of what happens during the process, it is sufficient
to explain exactly what happens in the process of thermonuclear fusion of
hydrogen into helium.
Every star contains both radiative and convective energy transfer regions.
The important difference between the two modes of energy transfer is that
radiative transfer regions do not move the material of the star, but convec-
tive ones do. This means that stars with large convective transfer regions
in their interiors may be able to fuse hydrogen longer than stars without
large convective transfer regions because those with large convective transfer
regions can move more “fuel” to the “fire.” That is, when the core becomes
depleted of hydrogen, a star that has a large convective transfer region near
the core can move more hydrogen into the core, thereby allowing the core to
fuse hydrogen for a longer period of time.
Interestingly, both extremely low- and extremely high-mass stars contain
convective transfer regions near the core. In high-mass stars about halfway
from the core to the surface, the energy transfer method goes to pure radiative
transfer, so about half the star’s material will never be able to reach the core.
Intermediate-mass stars, like the Sun, have large radiative transfer regions
near the core and transition to convective transfer regions at the outer layers.
In these stars, only the material that starts in the core will get fused. In low-
mass stars, the radiative transfer zone occurs only at the star’s surface. So all
the material in these stars will reach the core.
Another factor is the fusion process. In stars like the Sun and in lower-
mass stars, the fusion process is that described above. In high-mass stars,
however, the cores are generally hotter than in intermediate- and low-mass
stars. For this reason, another process in addition to the one described above
is used. That process is called the “CNO cycle” because it uses a carbon atom
as a catalyst to quickly use up hydrogen and make helium. In this cycle, a
carbon-12 atom fuses with a proton to make a nitrogen-13 atom, which
beta decays to a carbon-13 atom. The carbon-13 atom fuses with a proton
to make a nitrogen-14 atom. The nitrogen-14 atom fuses with another pro-
ton to make an oxygen-15 atom, which beta decays into a nitrogen-15 atom.
Another proton fuses with the nitrogen-15 atom making an oxygen-16 atom,
which emits a helium atom to become a carbon-12 atom again. The rates for
these reactions are much faster than the rates for the proton-proton chain,
and so these stars will use up their hydrogen and create a helium core much
more quickly than an intermediate-mass star. It is clear how the convective
transfer region occurring so close to the core is helpful in accelerating the
fusion reactions since it feeds the core with both hydrogen and carbon.
Taking all this information into consideration and knowing the reaction
rates for each reaction that occurs in the cores of main sequence stars, we
are able to determine a star’s time on the main sequence using the following
formula:
T = 1010 / M3.5
where T is the time a star will be a main sequence star in years and M is the
star’s mass in units of the Sun’s mass (or solar masses). (M3.5 is the same,
Putting Together the Puzzle of Stellar Evolution • 69
Already it is clear that what happens to a star must be somewhat tied to the
mass of the star. We have already seen that high-mass stars spend less time
on the main sequence than lower-mass stars because different processes
are going on inside them and because the conditions (temperature, pres-
sure, available atoms) in their cores are different. They have different ways
of distributing material to the core and different processes that go on inside
the core.
Intermediate- and low-mass stars have more in common, but there are still
some differences. In this section of this chapter we will talk about low-mass
stellar evolution. This route of evolution will be descriptive of what happens
to stars with masses less than eight times the mass of our Sun (but greater
than 0.4 times the mass of our Sun). For stars more massive than that, a dif-
ferent route of evolution will be followed. That route will be discussed in
the next section of this chapter. For stars less massive than 0.4 solar masses,
astronomers don’t know what happens. The first of these stars that formed
in the early universe have not yet stopped fusing hydrogen, so there is no
observational evidence of what they will become when they finish fusing
hydrogen.
The event that signals a change for a low-mass main sequence star is the end
of hydrogen fusion in the core of the star. This ceases because there is insuf-
ficient hydrogen to fuse into helium. Actually, the amount of hydrogen to
helium (by number of atoms) gets as low as 1 to 9 (that is 10% hydrogen by
number and 90% helium by number).
Because there is no significant amount of fusion occurring (there is, of
course, some fusion occurring, but it is not happening at an appreciable rate),
the core begins to contract. Gravitational collapse of the core sets in because
of the lack of radiation pressure from the diminished number of escaping
gamma ray photons. There is still gas pressure to resist the collapse, though,
so as long as the mass of the core is sufficiently high, it will not completely
collapse.
As the core collapses, the temperature of the core will rise and so will the
density of the gas within the core. The outer part of the star is doing some-
thing completely different. Fusion does not cease or diminish at all in a thin
70 • STARS AND GALAXIES
shell surrounding the core. As the core collapses, it takes some of the mate-
rial from the outer part of the star with it. That new material from the outer
part of the star condenses and gets hot enough to fuse hydrogen. That mate-
rial from outside the core is mostly hydrogen, so hydrogen is fused once
the temperature and density are high enough. This shell fusion (so-called
because it occurs in a shell of dense material surrounding the core) causes
something strange to happen to the outer parts of the star. The radiation
pressure from shell fusion pushes the outer parts of the star farther from the
core, making the star expand. At the same time, this material is getting less
dense and farther from the heating source, so it cools. The surface of the star
does exactly the opposite of the core. While the core contracts and heats up,
the outer parts expand and cool down.
Helium Flash
In the lower-mass stars (less than 2–3 solar masses) the core collapses,
but the pressure does not increase and the temperature does. This occurs
because the gas becomes degenerate. This means that there is no longer a
gas pressure associated with the material in the core, so as the temperature
rises, the pressure does not. In this case, the rising temperature of the core
causes the helium to be fused very quickly, creating what is known as a
helium flash.
A helium flash is not a brightening of the star, however, because this occurs
deep in the core of an expanding star. The energy released during the helium
flash is not manifested as radiation that escapes the star, but rather, works to
stabilize the core as radiation pressure, making the core no longer degener-
ate. Once the core has been stabilized, it begins to function like a normal
helium fusing core.
Electron Degeneracy
Electron degeneracy is the reason that the cores of some low-mass stars begin
to fuse helium in such an explosive way. This phenomenon is due to the fact
that the core is made of up ionized gases and free electrons. As the core col-
lapses, the gas becomes so compressed that it becomes degenerate. The free
electrons are pushed closely together, and a property called the Pauli Exclu-
sion Principle keeps them apart.
Electron degeneracy pressure occurs because no two electrons can have
the same properties at the same time. (This is the Pauli Exclusion Principle:
no two electrons can occupy the same quantum state at the same time.) Elec-
trons are structureless particles, unlike protons and neutrons. Like neutrons
and protons, in addition to their charge, electrons have properties like spin
Putting Together the Puzzle of Stellar Evolution • 71
and angular momentum. The sheer number of electrons in the cores of these
2–3 solar mass stars creates electron degeneracy pressure. This pressure is
constant and does not change with temperature (unlike gas pressure), so as
the core heats up, this pressure does not increase. As a result, helium fusion
begins and increases quickly.
In a normal (not degenerate) core, the increased temperature will increase
fusion rates, which will slow the contraction of the core as gas and radiation
pressure build up. In degenerate cores, the stopgap of increased gas pres-
sure does not occur, so fusion rates are not regulated and slowed by a small
increase in radiation pressure. Instead, fusion rates continue to increase in
a runaway reaction situation until enough radiation pressure is generated
to push the core outward against gravitational collapse and overcome the
degenerate state. Therefore, in degenerate cores, helium fusion does not start
gradually, but explosively.
The reason this occurs in lower-mass stars and not in higher-mass stars
is because in higher-mass stars the core is so big that the innermost part of
it can reach the temperature and density requirements for helium fusion to
occur without compressing the gas to degeneracy. This means that in the
cores of more massive low-mass stars (greater than 2–3 solar masses) helium
fusion begins gradually in the centermost part of the core and radiation pres-
sure builds up slowly so that the whole core is never compressed to the point
of degeneracy. Conversely, smaller cores cannot reach the temperature and
density requirements for helium fusion in their innermost parts. The entire
core of a lower-mass star must be compressed to degeneracy just to reach the
temperature requirement for helium fusion.
No matter the path, eventually stars that will fuse helium become red giants.
Once helium fusion has taken hold and stabilized the core of the star, the
outer parts of the star will begin to change. On the H-R Diagram, the star has
been traveling up and to the right along the red giant branch, increasing in
luminosity and size, while decreasing in surface temperature. When helium
fusion has stabilized the core, the star has reached the top of the red giant
branch. As the star continues its existence as a helium-fusing star, it migrates
back down and to the left on H-R Diagram along the horizontal branch.
This means the star is decreasing in size and luminosity as it increases in
temperature.
The helium fusion process is similar to the hydrogen fusion process, in
that it has many steps to it, produces new, heavier elements, and produces
energy. The helium fusion process is also known as the triple alpha pro-
cess. This name comes from the fact that helium atoms are also known as
alpha particles and each cycle includes three helium atoms, or three alpha
particles.
72 • STARS AND GALAXIES
The first step of the process is two helium-4 nuclei combine to form the
unstable isotope of beryllium called beryllium-8. This stage of the nuclear
reaction can be written as follows:
He + He → Be
Be + He → C + γ
C + He → O + γ
Again, no matter what the next phase will be, red giants that deplete their
helium cores do the same thing: they contract their cores further. Helium
and hydrogen fusion will continue in the outermost layers of the star, with
hydrogen fusion occurring furthest from the core. Ignition of these outer
layers will occur during the collapse of the core. Shell hydrogen and shell
helium fusion will have the same affect that shell hydrogen fusion did. The
star will expand due to the radiation pressure source being brought closer to
the outer layers, the increase in radiation pressure, will cause the outer layers
to expand and cool and the star will move up and to the right on the H-R
Diagram along the asymptotic giant branch.
This time, for all stars with masses less than about eight solar masses, the
core will contract until it is degenerate. This time, the degeneracy of the core
Putting Together the Puzzle of Stellar Evolution • 73
is what ignites the shell fusion layers. Eventually the expansion of the outer
layers of the star begins to drag the hydrogen fusion shell with it and this
shell becomes dormant. Next, the helium shell becomes dormant, then the
core turns off. Following this, the gas pressure in the shells decreases so that
they collapse and reignite. First, the hydrogen shell ignites and fuses hydro-
gen into helium. Then, a degenerate helium shell ignites in a flash not unlike
the helium flash of the stars with masses less than 2–3 solar masses. This flash
pushes out the hydrogen shell, causing it to go dormant again. This process
repeats itself many times during the asymptotic branch life of the red giant.
Each helium shell flash causes some of the outermost layers of the stars to
be expelled or shed from the star. Eventually, the star is just a degenerate
core with several expanding shells of material surrounding it. This is called a
planetary nebula.
A planetary nebula is the fate of all stars with masses less than about eight
solar masses. The name planetary nebula is perhaps a little misleading. Origi-
nally astronomers of the 19th century who discovered these objects specu-
lated that they may be solar systems forming. These astronomers had little to
go on but appearance at the time. (Perhaps the most famous planetary nebula
is the Ring Nebula, located in the constellation Lyra.) Now we know much
more about what these objects are and how they form than the astronomers
who named them. Nonetheless, the name has not been changed to reflect
better the nature of these objects.
A planetary nebula is the phase reached by a lower-mass star (between
about 0.4 and 8 solar masses) when it reaches the top of the asymptotic branch.
The appearance of a planetary nebula is an expanding ring or shell of hot gas.
Most of the radiation from the gas is from ionized hydrogen, but there is also
evidence for heavier elements (such as helium, carbon, and oxygen, notably).
The expansion rates of these gases are typically about 10 km/s to 30 km/s. No
planetary nebulae have been observed that are significantly older than about
50,000 years. It is thought that after such a long time, the gases have cooled
sufficiently that they do not radiate enough to be observed.
At the center of every planetary nebula is the cooling, degenerate core
of the original star. The core is roughly 40 percent or less of the mass of the
original star. The rest of the mass of the original star was ejected back into
space during the planetary nebula formation.
The hot degenerate core of the star will remain. This object is known as a
white dwarf. The name comes from its color (white), which is due to its very
74 • STARS AND GALAXIES
high temperature, and its size. White dwarf stars are small (about the size of
Earth) and very hot (from several hundred thousand degrees Kelvin down
to about 5,000 degrees Kelvin). They start off at the higher temperature and,
since they generate no energy within them, they cool off over time to tem-
peratures cooler than our Sun.
These objects are called “stars” but they have no method of energy produc-
tion like the other kinds of stars we’ve discussed. White dwarf stars produce
no energy. They are hot because they used to be fusion reactors. Now they
cannot even perform thermonuclear fusion because they are made of carbon
and oxygen atoms and are not hot enough to fuse these atoms to produce
energy. White dwarf stars will slowly cool over time.
Contrary to what might be expected, white dwarf stars do not change size
as they cool. In fact, they crystallize as they cool with the carbon and oxygen
atoms forming a lattice structure (“like a diamond in the sky”). The star just
cools and as it does so, fades away. With no change in size, the decrease in
temperature must decrease the luminosity. White dwarfs eventually become
black dwarfs (theoretical stars that emit no light).
In the previous section of this chapter, the evolution of stars with masses less
than eight times the mass of our Sun was described in great detail. In this
section of this chapter, the evolution of more massive stars will be discussed.
In the main sequence phase of the evolution of a massive star, things are already
different. The mass of the star is so great that it takes more outward radiation
pressure to keep these massive stars from collapsing on themselves. So, the
processes of thermonuclear fusion that power these stars are different.
Massive stars use the CNO cycle in addition to the process of thermo-
nuclear fusion of hydrogen described in the last section of this chapter. The
CNO cycle is a process whereby hydrogen is fused into helium using carbon,
nitrogen, and oxygen nuclei, which make up about 2 percent of the mate-
rial in a star (by number of atoms). Essentially, four hydrogen nuclei are
fused into one helium nucleus. This fusion of hydrogen in the CNO cycle
occurs in six steps where unstable isotopes of carbon, nitrogen, and oxygen
are formed. Because carbon works as a catalyst in this process, the hydrogen
is used up and helium is produced much faster than in the kind of ther-
monuclear fusion that occurs in lower-mass main sequence stars. The CNO
cycle requires a higher core temperature than the hydrogen fusion process
that goes on inside the cores of less massive stars. It also requires a higher
pressure and the presence of at least a critical amount of carbon atoms.
Stars with masses greater than eight solar masses have hotter cores and
greater pressures within those cores. The faster rate of fusion means these
stars will use up their hydrogen fuel faster than lower-mass stars. So, more
massive stars spend less time on the main sequence than less massive stars.
Putting Together the Puzzle of Stellar Evolution • 75
Just like lower-mass stars, massive stars transition to a helium fusion phase
after exhausting the hydrogen in their cores. This phase begins with hydro-
gen shell fusion turning on, which results in the expansion of the outer
shell material of the star, making the star a red giant. In the case of the most
massive stars, the size of the giant star is considerably larger than for lower-
mass stars. Astronomers call these most massive helium fusing objects red
supergiants.
Again, similar to the main sequence phase for massive stars, the red giant
phase is short-lived. Helium core fusion begins when the core collapses fur-
ther and reaches the critical temperature and density necessary for helium
fusion to begin. Just like hydrogen fusion begins in all stars, massive stars
begin helium fusion gradually, starting in the core where the temperature
and density are high enough first and moving outwards until the entire core
is actively fusing helium into carbon.
This phase of a massive star’s existence will last only a few hundred thousand
years, whereas this phase of lower-mass stars lasted millions of years. Follow-
ing the core becoming inert carbon, shell hydrogen and shell helium fusion
will begin outside the core, again causing the star to increase in size. The core
will undergo a small collapse causing it to heat up and become more dense.
Carbon Fusion
The result of this second collapse of the core of the massive star is carbon
fusion. Carbon fusion requires a core temperature of 600 million K. Carbon
fusion can be through helium capture (carbon fuses with helium to create
oxygen) or with other heavy elements (carbon fuses with oxygen to form
silicon). Carbon fusion will provide the necessary energy source to put the
star back into gravitational equilibrium until carbon fusion ceases, after a few
hundred years.
Following the carbon fusion phase, hydrogen, helium, and carbon shell
fusion will begin in the outer layers of the core and the core will collapse
again, igniting the oxygen ash and beginning the death sequence for the
massive star.
Fusion of heavy elements into heavier elements occurs for ever shorter and
shorter periods of time. These fusion processes are complex and can occur
simultaneously. The pattern of how these phases begin and end is the same,
however. Fusion occurs until the element being fused runs out, then fusion
stops. The outer shell layers of the core ignite when the core ceases fusion,
76 • STARS AND GALAXIES
the core collapses, heats up, and begins to fuse the ash left behind from the
previous reaction.
This can continue until the core becomes iron ash. Fusing iron nuclei
requires energy, so the star cannot create a new energy source by fusing iron.
Nor can the star produce energy by fission of the iron nuclei, as that also
requires an energy input. As a result, the star will not be able to return to
gravitational equilibrium. Instead, the star will collapse on itself in a cata-
strophic explosion called a supernova.
SUPERNOVAE
When a massive star runs out of elements that it can fuse in its core, it is no
longer in balance. While the star is producing energy in its core, it can bal-
ance, with an outward force, the force of the weight of the star so that the star
does not collapse. When the core of a massive star reaches the point that it
contains insufficient materials for fusion processes that produce energy, the
star collapses.
Supernovae
The name “supernova” comes from the fact that early astronomers who watched the skies for signs
from the heavens called anything that appeared in the sky that was not known before a “nova,”
which is Latin for “new.” The first supernova recorded by humans was recorded by Chinese and
Native Americans. Based on the expansion rate of the supernova remnant, the supernova recorded
is thought to be that which created the Crab Nebula. The only other supernova that occurred in our
galaxy, the Milky Way, known to be observed by humans is one observed by Tycho Brahe in 1632.
The word supernova has its origin in the early 20th century, according to the Oxford English Diction-
ary, however.
The outer layers of the star fall, unsupported, and crash into the iron core.
Depending on the size (or mass) of that iron core, one of two things will hap-
pen. The core may collapse, but remain supported against further collapse
by the force of repulsion between neutrons (neutron degeneracy), leaving
a remnant called a neutron star. Or, the core may be too massive to remain
supported by the strongest force in the universe and will collapse into a
black hole.
SUPERNOVA REMNANTS
When a massive star collapses in a supernova, the material of the star bounces
off of the inert iron core and spreads outwards from the remnant neutron
Putting Together the Puzzle of Stellar Evolution • 77
star or black hole for thousands of years after the explosive end of the massive
star’s existence.
The material is heated to extremely high temperatures and even more massive
elements are formed in the process, using some of the energy of the super-
nova event. The hot, low-density gas that was once the bulk of a massive star
expands into the near-vacuum of space interacting with whatever dust or gas
is in its path. The expanding shell of extremely hot, low-density gas can be
observed thousands to millions of years after the supernova event.
Such supernova remnants are often mistaken for other types of objects,
but fall into the category of “nebulae.” The most well-known supernova rem-
nants are several hundreds to several thousands of years old like the Crab
Nebula, the Veil Nebula, and the Cygnus Loop Nebula.
The two types of supernova remnant cores that can exist are neutron stars
and black holes. Neutron stars are not really stars, per se, as they don’t pro-
duce any energy in their cores. They are merely hot bodies that radiate their
heat and cool over time. A neutron star, similar to a white dwarf, is made of
inert material that is supported from collapse by the fact that the material is
also degenerate.
Whereas a white dwarf is supported by electron degeneracy pressure, a
neutron star is supported by neutron degeneracy pressure. Electron degen-
eracy pressure is a pressure or force due to the quantum rule that no two
electrons can occupy the same quantum space at the same time. That means
that no two electrons can have the same position, charge, spin, momentum,
and magnetic moment at the same time.
A good analogy of this “pressure” is to imagine how two people wearing
the same costume might behave if they appeared at the same costume party.
They would likely stay as far apart as possible at all times. This is how elec-
tron and neutron degeneracy pressure work to support these types of stars
against collapse. In white dwarf stars, the pressure is due to electron prop-
erties, whereas, in neutron stars, the pressure is due to neutron properties.
Neutrons have fewer quantum states (fewer possible properties in common)
than electrons, so neutron degeneracy pressure is stronger than electron
degeneracy pressure.
In fact, the force that governs neutron degeneracy pressure is so poorly
understood that astronomers can only estimate an approximate upper limit
for the mass of a neutron star. The best models suggest that a neutron star
with a mass greater than three times the mass of our Sun would not be able
78 • STARS AND GALAXIES
Black Holes
There is a common misconception that black holes are extremely massive objects, but actually, it is
not that they are massive, but that they are compact that they have such interesting properties. It is
the amount of mass in such a small space that makes them unique objects. For the Sun to become
a black hole, for example, it must shrink in size so that all its mass fit in a volume of radius 3 km.
(For comparison the radius of the Sun is currently 700,000 km.) Since the mass of the Sun does not
change in this scenario, the gravitational pull due to the Sun does not change either. So, if the Sun
were to become a black hole (which it cannot, since it is not massive enough to end up with an iron
core), the solar system would not be sucked into it.
Black holes do not radiate any electromagnetic energy, so how can astron-
omers tell they are there? While black holes don’t radiate electromagnetic
energy, they do have a gravitational field associated with them. This gravita-
tional field affects the way objects near the black hole will move. The more
massive the black hole the faster objects near it will move.
Putting Together the Puzzle of Stellar Evolution • 79
Also, some black holes attract material that is accreted onto the black hole
by way of an accretion disk of material. This disk of material gets heated
to extremely high temperatures and can radiate electromagnetic radiation
(since it is not part of the black hole). Sometimes, astronomers can identify
black holes by identifying an accretion disk.
T Tauri Phase
Low-mass stars (like the Sun) lose mass both at the beginning and end of
their existence. During the gravitational collapse of a cloud of interstellar
80 • STARS AND GALAXIES
medium during the formation of low-mass stars, the collapsing star is both
accreting mass from the collapsing cloud and ejecting mass into the collaps-
ing cloud. During its formation, the protostar spins faster as it collapses and
the material of the cloud forms a disk around the protostar. Material from the
disk accretes onto the surface of the star, and the star begins to expel mate-
rial in the form of jets that form at the poles along the axis of rotation of the
protostar.
The first such star observed was in the constellation Taurus. Until it was
studied well, it was not understood what the physical mechanism was that
caused its unusual properties of irregular variable brightness and emission
lines. Now, astronomers understand not only that there are many stars that
have the same unusual properties, but that all stars like our Sun go through
a phase in their evolution that causes them to exhibit these properties. The
material expelled by the jets of a star in the T Tauri phase includes both gas
and dust.
Protoplanetary Nebulae
Novae
When gas falls on a low-mass white dwarf but does not increase the mass of
the white dwarf to the point that it exceeds the Chandrasekhar Limit (1.4
times the mass of the Sun), the white dwarf undergoes a nova. A nova is a
brightening of a white dwarf due to fusion of hydrogen or helium occurring
on the surface of the star.
To create a nova event, material must be accreted onto the star. That
material will lie in a thin layer on the surface of the star. When the accreted
material reaches a critical density, fusion occurs simultaneously all over the
star throughout the entire layer of accreted material and the star appears to
brighten briefly (for several minutes to several days) until all fusible material
has been exhausted.
Putting Together the Puzzle of Stellar Evolution • 81
When all the fusible material has been fused, the star returns to its stable
state of inert material supported by electron degeneracy pressure. This period
of surface fusion results in a massive outflow of fusible material because
there are no outer layers of a star to press down and resist the radiation pres-
sure created by the epoch of massive fusion. The hot gas that escapes the
surface of the star becomes part of the interstellar medium again.
This process can repeat and continue to repeat until the mass accreted
by the white dwarf star causes its total mass to exceed a mass equivalent to
1.4 times the mass of our Sun. This mass of 1.4 times the mass of the Sun
is referred to as the Chandrasekhar Limit. If the mass of a white dwarf star
exceeds the Chandrasekhar Limit, the force of electron degeneracy pressure
will fail to prevent the star from total collapse. The result of this scenario is a
type of supernova that leaves behind no remnant.
Supernovae
Type I
considered standard candles and are used to measure the distance to distant
galaxies. Knowing how luminous the supernova is and how bright it appears,
astronomers can calculate the distance to the supernova.
Type II
The material that makes up the universe is recycled constantly through the
formation and evolution of stars. Clouds of atomic gas cool and condense,
forming cold, dense clouds of molecular gas. Within these clouds of cold gas,
stars can form. It takes just some push or pull or magnetic field to cause the
cloud to fall into gravitational collapse. Once the cloud is collapsing, the road
to becoming a star (or many stars) is an inevitable one.
The formation of a star involves many different stages. In the beginning
there is just dust and gas. Slowly, over time, the dust and gas form a cloud of
material containing all the right ingredients for star formation. Then, some-
thing happens to cause the cloud to gravitationally collapse. The cause can
be a nearby supernova, or stellar winds from a massive star going through
Putting Together the Puzzle of Stellar Evolution • 83
RECOMMENDED READINGS
Freedman, Roger, and William J. Kaufmann III. Universe. 8th ed. New York: W.H.
Freeman Company, 2008.
Hawking, Stephen. On the Shoulders of Giants. Philadelphia: Running Press, 2002.
Jastrow, Robert. Red Giants and White Dwarfs. New York: W.W. Norton and Com-
pany, 1990.
Seeds, Michael A. Astronomy: The Solar System and Beyond. 5th ed. Pacific Grove,
CA: Brooks Cole, 2006.
4
Variable Stars and Multiple
Star Systems
There is an entire class of stars that do not fit the working definition of stars
we’ve used throughout this volume. These stars are not in equilibrium. These
stars are in an extended period of transition. These stars are called “variable
stars.” They are important for many reasons. Some of them simply tell us
about the physical nature of variability in stars, while others can be used as
standard candles and can be used to determine the distance to objects in the
Galaxy or universe.
There are two basic types of variable stars: intrinsic and extrinsic. Intrinsic
variable stars are divided into four groups: pulsating, eruptive, cataclysmic,
and x-ray. Extrinsic variable stars are either eclipsing or rotating. Each of these
types of variable stars will be discussed in this chapter. Details about how, if
at all, the variable property of the star fits into the evolutionary sequence will
also be discussed.
These are stars whose brightness changes due to the star itself changing
in luminosity. An intrinsic variable star changes in luminosity because it
expands or contracts or because some internal process changes the energy
output of the star. In the category of intrinsic variable stars we include stars
that pulsate, stars that erupt, and stars that have outbursts on their surfaces
or accretion disks. This includes supernovae and novae (discussed in the
85
86 • STARS AND GALAXIES
previous chapter) as well as RR Lyrae and Cepheid variable stars, which are
used as distance indicators.
Stars become unstable when they make the transition from fusing one kind
of element in their cores to another. The most well-understood transition is
the transition from main sequence star to red giant, which signifies the tran-
sition from core hydrogen fusion to shell hydrogen fusion and core helium
fusion. All stars greater than about 0.4 solar masses go through this transi-
tion. On the H-R Diagram, this transition region is called the instability
strip.
The lower mass stars pass through this instability strip at a slower pace
than the higher mass stars. The lower the mass of the star, the longer the
stages of evolution. Inside the star, during this transition, the core of the star
ceases to fuse hydrogen and begins to gravitationally collapse. As it does, the
outer parts of the core get dense enough and hot enough to fuse hydrogen
and a process known as shell hydrogen fusion begins.
Shell hydrogen fusion causes the outer layers of the star to expand into a
red giant. Shell hydrogen fusion can also stop, as the thin shell of hydrogen
is quickly depleted, then re-start as more of the core material collapses,
drawing more of the shell material in closer to the core where it can fuse.
Core helium fusion starts up similarly. It can start, then stop and start
again.
When a star transitions to helium fusion in its core in this start-stop-start
manner, it goes in and out of stability, causing it to shrink and expand. Since
the material of the star does not escape it, the pulsations are regular as the
mass of the core slowly increases until it reaches a stable equilibrium point
where fusion can occur continuously, using up the fusible material in the
core. This means that some stars will become variable stars for part of their
existence as a kind of phase of their evolution. So, variable stars are not vari-
able all the time.
Variable Stars and Multiple Star Systems • 87
RR Lyrae Stars
One type of intrinsic pulsating variable star that is well-studied and under-
stood is the RR Lyrae variable star. The first star of this type was found in
the constellation Lyra. The star designated RR Lyrae is the first star to be
observed to have these properties. RR Lyrae stars are stars of spectral types
A–F that are on the lower portion of the instability strip found between the
main sequence and the red giant branch for these types of stars.
An RR Lyrae star has a very low metallicity (meaning a very small amount
of elements more massive than helium) and is usually found in a globular
cluster or other location typically populated by Population II stars. RR Lyrae
stars have a characteristic period of variation of that is usually 0.5 days with a
range of about 0.1 to 2 days. The amount of variation for these stars is usually
1 magnitude (with a range of about 0.3 to 2 magnitudes).
Because RR Lyrae stars are in such a specific part of the H-R Diagram,
the average absolute magnitudes of these types of stars is known. Therefore,
measuring a RR Lyrae’s average apparent magnitude allows one to calculate
the distance to an RR Lyrae star. This type of measurement (where the lumi-
nosity or absolute magnitude is known) is called a “standard candle mea-
surement.” It is as if you were looking at a series of lamps located at different
distances from you. Each one has the same type of lightbulb in it. So, sim-
ply observing which lamps appear brighter tells you which ones are closer.
Knowing exactly how much light you are receiving and exactly how much
light is emitted from each lamp, one can calculate the distance to the lamp.
Population II Stars
Population II stars are stars that occupy the spheroidal components of our Galaxy (i.e., the bulge
and the halo). Due to their low content of elements heavier than helium (low metallicity), they are
thought to be stars that formed longer ago than stars that contain more elements heavier than helium.
Since all elements heavier than helium were formed in stars either through fusion in their cores
or through fusion processes that occurred during their cataclysmic ends, the presence of elements
heavier than helium indicates that these stars formed in an interstellar medium enriched by the recy-
cling of materials through the evolution of a previous generation of stars. So, low metallicity stars are
necessarily older than high metallicity stars. The implications of this information on how the Galaxy
must have formed are discussed in detail in chapter 6.
Another type of intrinsic pulsating variable star is the Cepheid variable star.
These stars are known to be yellow giant or supergiant stars. On the H-R Dia-
gram they are found in the upper portion of the instability strip. These stars
are subdivided into two groups Type I and Type II.
88 • STARS AND GALAXIES
The Type I Cepheid variable stars are also known a Delta Cephei stars.
The first object discovered to exhibit these properties was Delta Cephei, the
fourth brightest star in the constellation Cepheus. These stars are Population
I stars found in the spiral arms or disk region of the Galaxy. Delta Cephei
stars have periods of variation that range from about 5 to 10 days. Delta
Cephei stars are found on the upper right part of the instability strip, as they
are more luminous than the Type II Cepheid variable stars.
Population I Stars
Population I stars are stars with high metallicity. That is, they contain elements in their spectra that
indicate that they were formed from an interstellar medium that had been enriched with elements
heavier than helium. In other words, Population I stars are formed from recently recycled material.
Therefore, they are necessarily younger than Population II stars.
The Type II Cepheid variable stars are also known as W Virginis variable
stars. The first object to be discovered to exhibit these properties was the star
in the constellation Virgo named W Virginis. The Type II Cepheid variable
stars are distinctly different from the Type I stars because they are Popula-
tion II stars found in the spheroidal parts of the Galaxy. W Virginis stars have
periods of variation that range from about 10 to 30 days.
Both of these types of Cepheid variable stars were, and still are, used to
measure distances. Cepheid variable stars have the unique property that their
period of variation in luminosity is related to their actual average luminos-
ity. That is, the longer the period of variation, the brighter the star. For this
reason, they are used as (varying) standard candles. To get the distance to a
globular cluster or galaxy containing a Cepheid variable, measure its period
of variation and its average apparent magnitude. Knowing the period allows
one to determine the average luminosity, or average absolute magnitude
using the period-luminosity law. Knowing both apparent and absolute mag-
nitude, the distance can be calculated using the distance modulus formula:
m − M = 5 log D − 5
The ZZ Ceti star is the only type of non-radial pulsating intrinsically variable
star discussed in this volume. A ZZ Ceti star is a white dwarf star that changes
in brightness only slightly (0.001 to 0.2 magnitudes) over very short peri-
ods of time (30 seconds to 25 minutes). The first star discovered to exhibit
this type of behavior was the star named ZZ in the constellation Cetus.
There are three different types of this kind of variable star, distinguishable
only by their spectra. The three types reflect the temperature and composi-
tion of the white dwarf star. The first is DA for a white dwarf star with a
hydrogen-rich outer layer. A DA star is distinguished by strong hydrogen
absorption lines in the spectrum. The next is DB for a white dwarf star with a
helium-rich outer layer. A DB star is distinguished by strong neutral helium
absorption lines. Finally, a DO white dwarf star is one with a helium-rich
outer layer that is distinguished by ionized helium absorption line, indicating
that the star is much hotter than the DB type star.
The short period and slight changes in brightness, as well as the fact that
usually several periods of variation are observed at once indicate that these ZZ
Ceti stars are not radially pulsating. Rather, these stars are experiencing non-
radial pulsations that are warping the surface of the star asymmetrically.
Eruptive or cataclysmic variable stars are stars that have occasional violent
outbursts caused by activity in its chromosphere or corona or enhanced stel-
lar wind or coronal mass ejections. A nova and a supernova would be con-
sidered types of cataclysmic variables. Most of the types of eruptive variable
stars discussed in this chapter are repeating variable stars, unlike supernovae,
but they are also mostly irregular variables as their variation in luminosity is
usually due to a physical process that is irregular.
T Tauri stars were discussed briefly in the last chapter in the context of recy-
cling matter in the universe. This type of variable star is an irregular vari-
able star, which means that there is not really a period of variation range or
brightness variation range. These stars are identified by their spectra. They
have the spectra of low-temperature (G–M spectral class) stars with strong
emission and absorption lines. In particular, lithium is observed in the spec-
tra of these stars, indicating that they are very young objects (probably less
than 10 million years old).
Another indication of the youth of these objects, is the fact that they are
enshrouded in an envelope of dust and gas and have large protoplanetary
Variable Stars and Multiple Star Systems • 91
accretion disks. The luminosity of a T Tauri star changes, most likely, due to
instabilities in the disk, violent activity in the atmosphere of the star or mov-
ing clouds of dust and gas. T Tauri stars are an early phase in the evolution of
low-mass stars, like our Sun.
UV Ceti stars are red or orange low-mass dwarf stars. These stars exhibit
rapid, irregular variations in brightness that are not predictable. UV Ceti
stars are also known as flare stars. Their luminosities can change abruptly in
many different parts of the electromagnetic spectrum at once or in sequence.
The spectra of UV Ceti variable stars include hydrogen and calcium lines,
indicating activity in the chromospheres of these stars.
Astronomers equate the activity in UV Ceti to the activity we observe
on our own Sun. On the Sun, flares are caused by magnetic energy being
released through a rapid heating of plasma, which manifests itself as a spike
in brightness of a portion of the solar material. In UV Ceti variable stars,
the flares must be significantly larger (compared to the star from which the
flare originates, since most UV Ceti stars are cool K–M spectral type dwarf
stars) and tend to cause spikes in the ultraviolet-blue portion of the electro-
magnetic spectrum.
merging of a binary white dwarf system or a final helium shell flash where a
pre-white dwarf star inflates its remaining small envelope resulting in a “re-
born” asymptotic giant branch star. Either way, however, this phase is associ-
ated with medium- to low-mass stars.
The prototype star of this class of stars is found in the constellation Orion. Its
unusual behavior that defines it as a type of variable star is that it brightened
by 6 magnitudes (100 times its original brightness) over a period of about a
year. During this brightening phase, a nebulous glow was observed surround-
ing the star. The brightening of the star was not a smooth uniform brighten-
ing, but rather it was due to multiple bursts of increasing brightness that do
not diminish. The star’s brightness no longer varies, and there are now several
(more than 10) examples of other stars that have exhibited similar behavior.
These stars are pre-main sequence stars in early stages of their develop-
ment. The observed increase in brightness is thought to be caused by thermal
instabilities that arise in the inner portions of the accretion disk surrounding
the star. The thermal instability causes the gas in the disk to become ionized,
and releases a burst of radiation when the hot material is accreted onto the
star. The amount of matter accreted determines the length and brightness
of the burst, so it is not possible to predict the length or duration of the
brightening.
Although it has not yet been observed, current theories suggest that this
epoch of brightening from accretion of hot material is part of a repetitive
process. The cycle may take much longer than a human lifetime, with long
periods of relative quiescence (inactivity). Astronomers are still studying FU
Orionis in order to learn more about what caused its brightening.
Wolf-Rayet Stars
Wolf-Rayet stars are a class of stars that are identified by their spectra, rather
than their variability. They are variable stars, and their variation in luminos-
ity is irregular. Wolf-Rayet stars are hot, massive stars in an advanced stage of
evolution. During this stage of evolution, the massive stars are expelling mate-
rial in the form of massive stellar winds. Wolf-Rayet stars are thought to be O
stars that have had their outer hydrogen envelopes stripped due to their being
a member of a close binary system (such systems are described in detail in the
next section of this chapter). The massive helium core is exposed. Depending
on the initial mass of the star, the core can include other elements.
There are two subclasses of Wolf-Rayet stars that can be identified by their
spectra. The type known as WN have prominent emission lines of helium
and nitrogen. The type known as WC have prominent carbon, oxygen, and
Variable Stars and Multiple Star Systems • 93
helium lines. The difference between these two types is most likely due to
the initial mass of the star in question. More massive (M > 25 M) stars
may reach this phase with heavier elements formed since they may be fus-
ing hydrogen using the CNO cycle, rather than the proton–proton chain,
whereas less massive stars (9 M< M < 25 M) may not have as many heavy
elements formed when they reach this stage in their evolution.
U Geminorum
This class of variable stars is defined by their irregular and extremely bright
bursts of luminosity. These stars are dwarf stars that are in a close binary
star system. The stars in this binary system are so close that mass transfer is
occurring between the stars. The white dwarf star is accreting matter from its
binary partner. As the material leaves the partner star, it accumulates in an
accretion disk around the dwarf star. Periodically, an instability occurs in the
disk (or in the other star) that causes the rate of accretion of material onto the
dwarf to increase significantly. This increased rate of accretion results in an
explosive outburst, which brightens the star for a short period of time.
The brightening in this case is due to the same phenomenon that causes
a nova. Material accumulates on the surface of the white dwarf, causing
the white dwarf to shrink and condense. When the density of the material
accreted onto the star is high enough for fusion to occur, it does so all at once.
This creates a very short, very bright burst in energy output from the star.
Symbiotic Stars
Symbiotic stars are star systems where two stars are gravitationally bound.
The typical components of a symbiotic star system are a red giant and a white
dwarf. The two stars are usually surrounded by nebulosity that is thought
to be material from the red giant that is being lost through stellar winds or
pulsation. The variability observed in these systems is caused by mass loss
from the red giant to the white dwarf resulting in a nova. In this system,
however, the material being lost from the red giant is not accreted onto the
white dwarf through an accretion disk. In fact, the two members appear to
be detached. No accretion disks have been observed in these systems. Rather,
the material is thought to fall on the surface of the white dwarf.
The result of this activity is that these stars experience irregular variations
in brightness. The change in brightness can be as much as 4 magnitudes. The
frequency of these variations is not regular and is separated by several hun-
dred days. This class of objects is not very homogeneous; no two objects in
this category share all their characteristics.
This type of variable star is also a binary system, but in this case the binary
system includes a neutron star or black hole in addition to another star. These
systems are characterized by the fact that they emit x-rays. The x-rays come
from material in the accretion disk surrounding the compact object being
heated to extremely high temperatures. If the star in the system is like our
Sun, the system is classified as a low-mass x-ray binary (LMXB); if the star
is greater than 10 M, the system is classified as a high-mass x-ray binary
(HMXB). This class of star experiences irregular variations in brightness
with no limit to the range of brightness changes or period length.
An extrinsic variable star is one whose brightness fluctuates for reasons unre-
lated to the physical nature of the star. There are two basic types of stars that
experience such changes in brightness. There are eclipsing binary stars and
rotating variable stars.
Eclipsing variable stars are stars that are part of a binary system where one
star passes in front of another, as seen from Earth. When the hotter star is
Variable Stars and Multiple Star Systems • 95
eclipsed by its cooler companion, the total luminosity of the system decreases.
Eclipsing variable stars have very distinctive light curves with flat-bottomed
minima and maxima that are separated by very regular intervals of time. The
luminosity changes vary from system to system and depend on both the sur-
face temperatures of the stars in the system as well as the distance between
the two stars. Eclipses can occur in cataclysmic binary systems as well.
Algol
Algol type stars are eclipsing binary stars that form a close binary (where
mass transfer is possible), but are detached or semidetached. They have peri-
ods of 5 hours to 30 years and experience brightness variations of several
magnitudes. If mass transfer takes place, it is by direct accretion, rather than
accretion disk, as in symbiotic variable stars.
Beta Lyrae
This type of system is a close binary star system that is both an ellipsoidal vari-
able and an eclipsing binary. An ellipsoidal variable star is one whose shape
has been distorted due to gravitational attraction to its companion so that as
it rotates, its surface area, as seen from Earth, varies, causing its brightness
to vary. The variation of brightness in a Beta Lyrae type of variable is small
(a few magnitudes) and smooth, though the variation is related to both the
eclipsing and ellipsoidal properties of the system. While the circumstances of
this type of variable seem unusual, Beta Lyrae is thought to be an example of
one phase in the evolution of certain close binary systems.
W Ursae Majoris
This type of variable star is a more extreme type of binary system than the
previous two. In this case, there is a binary system, but the stars in it are so
close, they are almost touching. This type of binary system is called a contact
binary. Since the stars are so close, this type of variable star is characterized
by its short period and similar change in brightness (between the primary
and secondary minima).
brightness of the star varies as seen from Earth since different amounts of the
surface at its normal brightness are seen at different times. These variations
are regular since they are related to a feature on the surface of the star that is
relatively constant.
Ellipsoidal Variable
BY Draconis
The BY Draconis type of variable star is a star with starspots (similar to sun-
spots, but on other stars) that causes its brightness to vary slightly (from a
few 0.01 magnitudes to 0.5 magnitudes). These stars are usually spectral type
G, K, or M, so they can be dwarf stars, or stars like our Sun. The periods
range from a few hours to over 100 days. This type of variable can also be
variable in brightness due to flares on the surface of the star.
Although these stars vary in brightness for intrinsic reasons, they fall
under the category of rotating stars, which, in its most common form, is an
extrinsic variable. However, these stars are intrinsically variable stars whose
variation is due to the star’s rotation.
FK Comae Berenices
SX Arietis
This type of star is a main sequence B star with emission lines (helium and
silicon) that varies in brightness. These stars also have strong magnetic fields.
Variable Stars and Multiple Star Systems • 97
The variation in brightness of the star is typically about one day, which is
about the same as the rotation period of the star. The variation in brightness
observed is usually about 0.1 magnitudes.
Binary stars and multiple-star systems are common in the universe. The
process of star formation forms at least as many multiple-star systems as it
does single stars, like our Sun. Like binary star systems, in multiple-star sys-
tems the stars orbit a common center of mass (called the barycenter). This
can be observed in nearby systems of stars where both radial and transverse
motions can be measured. To determine whether a system of stars is a binary
or multiple-star system, it is necessary to observe the stars for a long enough
time to verify that they are, indeed, gravitationally bound to one another.
Our Sun is somewhat unusual, then. Not only because we live on a planet
that orbits it, but also because it is a single star, with no other star gravitation-
ally bound to it. Our Sun is not part of a binary or multiple-star system, like
most of the stars in our Galaxy, and, presumably, in the universe. It is unusual
to find a star that is not gravitationally bound to another star, but this is not
a rare type of star, just not the most common type of star.
Of course, it could be that our Sun was a member of a multiple-star system
and got kicked out. (That happens sometimes, but because of gravitational
interactions, not because the other stars don’t like it!) But, the characteristics
of the velocity (speed and direction) with which our Sun travels around the
Galaxy do not support this as a possible scenario. All the evidence points to
a lonely existence for the Sun with no gravitationally bound partner for the
Sun at any time in its existence.
than the planets, Sun, and Moon, so models of the universe showed only the locations of the planets,
Sun, and Moon relative to Earth.) That turned out to be incorrect. The Sun is at a focal point of each
of the elliptical orbits of the each planet in the solar system.
After that, astronomers thought the Sun was located at the center of the universe or galaxy. (These
astronomers did not know yet that the universe consisted of many billions of galaxies; they thought
the Milky Way was the universe.) Heliocentrism (the belief that the Sun was the center of the uni-
verse) turned out to be incorrect. The Sun is just one of hundreds of billions of stars in our galaxy,
the Milky Way. It is located in the disk of our spiral galaxy about halfway out from the center of the
galaxy.
So, being burned two times, astronomers are loathe to hold the position that Earth and the Sun
are somehow so special that there is nothing like them in the universe. Most astronomers would
agree that it is possible that this is true, but not likely, given what we know about the universe.
Recent studies of exoplanets indicate that planets can form in multiple-star systems. Several of the
known planets that orbit stars other than our Sun are orbiting stars that have gravitationally bound
companions. The state of at least a few of these systems indicates further that planets can even survive
post-main sequence evolution of nearby companion stars (which is discussed in great detail in this
chapter). It appears that the formation of planets is not an unusual process at all.
The fact that no astronomer has yet detected evidence of an Earth-like planet in a multiple-planet
system (like our solar system) does seem to support the anthropocentric point of view; however,
astronomers who study exoplanets insist that the lack of evidence is due to their inferior ability to
detect such small planets. It is true that planet finders are still working to enhance their methods
for planet detection to be able to find smaller planets. Most exoplanets found to date are more like
Jupiter than like Earth.
Binary Systems
Binary star systems are very common. As we discussed in the previous sec-
tion of this chapter, there are many types of variable stars that are binary sys-
tems. This is the simplest gravitational interaction one can describe between
stars: two stars rotating around a common point that is the center of mass of
the system.
Simple as it may seem, there are still many kinds of binary star systems
that astronomers have studied over the years. In large part, these “types” have
to do with how they are observed, and not some physical property of the
systems that makes them unique.
Visual Binaries
The first type of binary star system is the visual binary star system. This is a
pair of stars that appear to be orbiting one another. This type of binary star
system may or may not be physical. That is, the stars may be gravitationally
bound, or not. If the stars are not gravitationally bound and do not orbit one
another, the pair is referred to as a “double star” to indicate the lack of physi-
cal connection between the stars.
Variable Stars and Multiple Star Systems • 99
Astrometric Binaries
Physical binaries may include a star that is too faint to be detected. Alterna-
tively, the binary system may include a compact object, like a black hole, that
does not radiate any electromagnetic energy. In such cases, it is still possible
to tell that there are two masses orbiting a common center of mass. If only
one star is visible, careful observations of the star’s position can reveal an
invisible partner.
Because both stars are orbiting a common center of mass, both stars are
moving. No matter at what angle this system is observed from Earth, there
will be transverse motion of the visible star, as long as the system is close enough
to Earth to measure such motion.
The process of careful measurement of positions of stars is called astrom-
etry. Since these systems of stars can only be identified by such measure-
ments, they are called astrometric binary star systems. Since both members
are orbiting a common point in space (their center of mass), this type of star
system is different from visual binaries, as all astrometric binaries are physi-
cal, whereas only some visual binaries are physical. Additionally, in visual
binary systems, two stars are observed; whereas, in astrometric binary sys-
tems only one star is observed.
An example of an astrometric binary system is Sirius A and B. Only
using special instruments called stellar coronographs can the companion
100 • STARS AND GALAXIES
Figure 4.2The above drawing shows how both stars in a binary system move around
a common center, rather than the common idea of one star moving around another.
[Jeff Dixon]
star, Sirius B, be observed. Sirius is the brightest star in the southern hemi-
sphere. It is one of the bright stars that make up the asterism called the Win-
ter Ellipse. Looking south, Sirius is the star at the lowest point in the Winter
Ellipse. It is the brightest star in the constellation Canis Major (also known
as the Big Dog).
The presence of the companion star, Sirius B, was inferred from the care-
ful measurement of the position of Sirius A over many decades. The primary
star seems to “wobble” with regularity. This is the signature of an astrometric
binary star system. Not long ago, with the Hubble Space Telescope some of
the first images of Sirius B were taken.
Spectroscopic Binaries
a binary system could have that would not include any radial motion as seen
from Earth.
However, for those binary systems that are oriented so that at least a por-
tion of their motion around their center of mass is radial motion as seen
from Earth, astronomers can measure their radial motion and identify that
the star system is a member of a binary system. The radial motion is detected
because of the Doppler Effect, which causes a Doppler shift in the wave-
length of light observed. Just as the pitch of the sound of a train changes as
it passes you from high pitch to low pitch, the light emitted from an object
moving away from Earth is shifted to longer wavelengths. Conversely, the
light emitted from an object moving towards Earth is shifted to shorter wave-
lengths. So, to distinguish this system as a binary star system, astronomers
have to observe its spectrum.
For this reason, these systems are called spectroscopic binary star systems.
A spectroscopic binary is one that cannot be observed visually or measured
astrometrically. A spectroscopic binary is a system consisting of two stars
that are gravitationally bound to one another and orbit a common center of
mass. The only way to determine that the single star that is observed is actu-
ally two stars that are physically connected to one another is to observe and
analyze its spectrum. There are two different kinds of spectroscopic binaries:
single line and double line.
Eclipsing Binaries
This configuration of binary stars was briefly discussed in the previous sec-
tion of this chapter. Eclipsing binary stars are stars that are in a gravitationally
bound system and orbit a common center of mass and whose alignment with
respect to Earth is nearly edge-on so that one star passes in front of the other
as seen on Earth. When the dimmer (cooler) star is in front, the brightness
of the pair decreases most (primary minimum). When the brighter (hotter)
star is in front, the brightness of the pair decreases the least (secondary mini-
mum). When both stars are visible from Earth, the brightness of the pair is
at its maximum.
Eclipsing binaries are very important in astronomy because they are the
only systems that allow for the measurement of a star’s mass and radius. A
visual astrometric binary system will allow for a measurement of each star’s
mass. Spectroscopic binary systems allow for the direct measurement of the
speed of the stars when they are moving radially with respect to Earth. But,
without knowledge of the separation between the stars, the inclination of the
system with respect to Earth, or the distance to the stars in question, only
the mass of the system (not the individual stars) can be inferred from that
information.
With eclipsing binary systems, the inclination of the system is limited to
a very narrow range, so estimates can be made without knowing the exact
inclination. Hence, eclipsing binary systems that are not also visual astromet-
ric or spectroscopic binaries are still useful. However, to determine a mass of
an eclipsing binary system, the distance to the system must be known. The
speed of the stars and the diameter of each star are measured from the light
curve.
Figure 4.3 The above diagram shows how the brightness of an eclipsing binary star system
changes as the two stars pass in front of one another. When the brighter star is in front
of the fainter star, the peak is not as low as when the fainter star eclipses the brighter star.
When both stars are visible, the brightness is at a maximum for the system. [Jeff Dixon]
Variable Stars and Multiple Star Systems • 103
The primary minimum is the deeper dip in the curve; this is when the
cooler (dimmer) star is in front of the hotter (brighter) star. A portion of the
light coming from the brighter star does not reach Earth, instead, the light
coming from the dimmer star does. The secondary minimum is the shal-
lower dip in the curve; this is when the hotter (brighter) star is in front of
the cooler (dimmer) star. In this case, the light coming from the dimmer star
does not reach Earth, instead the light coming from the brighter star does.
The brightness maximum is the sum of the brightnesses of the two stars since
light from both the dimmer and the brighter star is reaching Earth.
Since the x-axis of this curve is time, it is possible to measure the time it
takes for one orbit to occur. Knowing the distance to the system and assum-
ing circular orbits, it is possible to determine the ratio of the masses of the two
stars. (If the angular separation of the stars is measurable, the mass of each
star can be determined.) Using the light curve, knowing the time it takes for
the dimmer star to pass in front of the brighter star, and vice versa allow for
the calculation of the linear diameter of each star. (If the light curve shows
flat-bottomed minima, then the actual diameter of each star is measurable,
otherwise the distance measured is only some fraction of the diameter of each
star. In other words, a minimum or lower limit to the diameter is measured.)
Multiple-Star Systems
In some star systems, at least two stars are close enough to one another that
when one expands to become a red giant or supergiant, some mass can be
transferred to the other. To understand how this can happen, we must first
understand something about the gravitational field between these two orbit-
ing bodies. It is, perhaps, easy to think about the gravitational attraction
between two bodies along a line that connects them.
Assuming we are talking about two stars with the same mass, as we go
along this line from one star to the other, the gravitational pull of the nearer
star dominates. In this case, at a point exactly between the two stars, the pull
of each is the same. At this point, a test mass would feel a net force of zero
because both stars would be pulling equally on that mass in exactly opposite
directions. This point is called a stable point. An object could remain at this
point in space (exactly half way between two equal masses) indefinitely. If the
object got just a smidge closer to either star, it would be pulled toward that
star, however. This kind of stable point, where a small perturbation results
in an object losing stability, is called a “quasi-stable point.” It is sort of like a
spherical rock balanced on the peak of a hill. Any push in any direction will
send the rock down the hill.
Now, consider a system that is more realistic: the masses of the two stars
are not equal. In this case, the stable point will not be exactly half way between
the two stars, but somewhere else. Where? Well, if we consider that the more
massive object exerts more force on a test mass over a greater distance, then
the point of stability must be closer to the lower mass object. This can actu-
ally be calculated using the Universal Law of Gravitation:
F = Gm1m2/r2
RECOMMENDED READINGS
DeGrasse Tyson, Neil, Charles Tsun-Chu Liu, and Robert Irion. One Universe: At
Home in the Cosmos. Washington, DC: Joseph Henry Press, 1999.
Freedman, Roger, and William J. Kaufmann III. Universe. 8th ed. New York: W.H.
Freeman Company, 2008.
Hawking, Stephen. On the Shoulders of Giants. Philadelphia: Running Press, 2002.
Jastrow, Robert. Red Giants and White Dwarfs. New York: W.W. Norton and
Company, 1990.
Seeds, Michael A. Astronomy: The Solar System and Beyond. 5th ed. Pacific Grove,
CA: Brooks Cole, 2006.
5
What Makes a Galaxy?
At some point in your life, if you haven’t done it already, you should go to a
dark location (far from city lights) on a clear summer night and look at the
Milky Way. Immediately, you will understand why it is named the Milky Way,
since it looks as if someone poured a large container of sparkly milk across
the night sky. That’s our Galaxy, the Milky Way. In fact, the word “galaxy”
comes from the Greek word for the Milky Way ( galaxias), which is derived
from the words for mother’s milk ( gala) and circle (kuklos). The Greek mean-
ing has to do with a myth about how the Milky Way was formed.
In 1609, Galileo was the first person to record an observation of the Milky
Way showing that it was made up of many stars. Before this observation,
it was believed that the Milky Way was a single cloud-like structure that
spanned the heavens. There was speculation that it might be made of gases.
No one, however, thought that it might actually be billions of stars.
So, is a galaxy billions of stars? Well, it’s a little more than just stars. A gal-
axy contains stars, gas, and dust. All these materials must be gravitationally
bound together to make up a galaxy. The remainder of this chapter will dis-
cuss the constituents of a galaxy: gas, dust, and stars, and their many forms
within a galaxy.
GAS IN GALAXIES
The gas found in galaxies is nothing like the gas you put in your car; it is
more like the air you breathe. We call the gas found in galaxies interstellar
gas. Interstellar just means between the stars. Gas (along with dust) in galax-
ies is found in between the stars. In the case of interstellar gas, the word “gas”
107
108 • STARS AND GALAXIES
States of Matter
Matter comes in four known states: solid, liquid, gas, and plasma. Most of us have significant expe-
rience with the solid and liquid states of most matter. Some of us have significant experience with
gaseous matter (beyond just breathing it!). Almost none of us have significant experience with the
plasma state of matter, but it is most important in astronomy as much of the hot gas in outer space
is in the plasma state.
Primarily, neutral atomic interstellar gas is hydrogen. Since the gas is neutral,
this means that each hydrogen atom consists of one proton and one electron.
So, each atom has a net charge of zero (since the proton has a positive charge
of one and the electron has a negative charge of one).
Typically, in astronomy, the Bohr model of the atom is used as a refer-
ence. The Bohr model describes an atom as a nucleus with electrons in orbit
around the nucleus. The orbits in the Bohr model are circular. This is where
the model is inaccurate—actually, the electrons move in quasi-chaotic orbits
around the nucleus and the orbits have several different shapes that define
them. For our purposes, though, the Bohr model is complete enough. The
shape of the orbits is irrelevant to our discussion. Rather, what is impor-
tant is that each orbit represents a different energy state for the electron. In
this aspect, the Bohr model is consistent with modern atomic theory. Fig-
ure 1.12 in chapter 1 of this volume shows a diagram of the Bohr model of
a hydrogen atom.
Each energy level in the atom, as defined in the Bohr model, defines an
“allowed” state for the electrons in that atom. In the case of hydrogen, there
is only one electron, but the different orbits still define energy states in which
that electron can exist. In a hydrogen atom, the electron cannot have an
energy that is in between any two of the energy states.
When hydrogen is in the neutral form, the electron is usually at its low-
est energy state (or ground state). That means the electron is in the smallest
orbit around the proton. Neutral atomic hydrogen has two ground states.
This is because both the electron and the proton are spinning. Since both
components of the atom are spinning, as well as orbiting one another, there
are two possible ground states. The higher energy ground state occurs when
What Makes a Galaxy? • 109
the spin of the electron and the spin of the proton are in the same sense
(both particles spinning clockwise or counterclockwise). The lower energy
ground state occurs when the spin of the proton and the spin of the electron
are in opposite directions (one spinning clockwise and the other spinning
counterclockwise). The transition from the higher energy ground state to the
lower energy ground state is called a spin-flip transition, since the electron
changes orientation so that its spin is opposite that of the proton.
The spin-flip transition is a change in energy equivalent to the energy of a
photon with wavelength 21cm. Neutral atomic hydrogen gas, therefore, emits
electromagnetic radiation at a wavelength of 21cm. Through observations of
this emission line, astronomers can study neutral atomic hydrogen. Observa-
tions of neutral atomic hydrogen in the Milky Way tell us about the rotation
of the Galaxy. Observations of 21-cm radiation in other galaxies tell us about
the mass contained within those galaxies as well as the rotational motion of
those galaxies (galaxy rotation and how it is related to mass are discussed in
chapter 7 of this volume).
of gas within an accretion disk being consumed by the black hole. These
incidents of mechanical (sometimes called collisional) processes that ionize
interstellar gas are usually indicative of unusual physical conditions.
Ionized gas is most often studied by observing the emission spectrum
of the gas. This has historically been done primarily at visible wavelengths.
However, in the last few decades new instruments have been developed to
detect radiation in the ultraviolet, infrared, and microwave regimes; these
instruments make it possible for astronomers to study more of the emission
spectra of astronomical objects than ever before. As a result, astronomers
are learning more about the kinds of ionization processes that go on in the
interstellar medium.
Plasmas can be created in laboratory situations and studied using at the
same wavelengths that astronomers study the interstellar ionized gas. Based
on the laboratory studies, astronomers can learn about the density and tem-
perature of interstellar ionized gas. In the laboratory studies, different transi-
tions of the electrons within an ionized atom will be more or less frequent
depending on the ionization or excitation method and the density and tem-
perature of the ionized gas. So, astronomers can take the laboratory data,
as well as the theoretical predictions for scenarios that cannot be recreated
in a laboratory, and compare these to what they observe in the interstellar
medium.
Interstellar gas in molecular form is mostly molecular hydrogen gas (H2). The
second-most abundant and well-studied interstellar molecular gas is carbon
monoxide (CO). There are plenty of other, more complex, molecular gases
found in the interstellar medium, but they are far less abundant and remain
difficult to understand.
Molecular hydrogen, (H2) as it turns out, cannot be studied directly since it
does not radiate significantly at any observable wavelength. (Except for a col-
lisionally excited emission line found in the near-infrared regime, there is no
well-studied emission from H2.) This has to do with the structure of H2. Two
hydrogen atoms joined together have no dipole moment. That is, the mol-
ecule has no preferred orientation in the presence of an electric, magnetic,
or gravitational field. Therefore, the presence of any of these fields will not
cause the atom to lose an electron, rotate, or vibrate, so the electrons will not
change energy states and no electromagnetic radiation will be emitted. Emis-
sion from H2 will only occur if the hydrogen molecule is caused to vibrate or
rotate by some mechanical process. This appears to happen only in extreme
situations, like near supermassive black holes or recent supernova events.
To study H2, astronomers have studied extensively CO emission in the
radio regime. CO seems to always be associated with molecular hydrogen.
Astronomers have even tried to derive a relationship between the amount
What Makes a Galaxy? • 111
CO detected and the amount of H2 associated with the CO. The relationship
is still not one that is widely agreed upon. Different astronomers use different
methods to derive the relationship, and many different answers are consid-
ered acceptable. (Much like the value of the Hubble Constant in cosmology.
For more information on the Hubble Constant and its use in cosmology, see
the Cosmology and the Evolution of the Universe volume of this series.)
The primary method for the production of electromagnetic radiation from
molecular gas is through rotation or vibration of the molecule. Since both of
these processes involve accelerating a charge or group of charges, electromag-
netic radiation is produced. These processes involve very low energies, so the
radiation produced is often at very long wavelengths. Interstellar molecular
gas, therefore, is primarily studied in the radio regime.
Molecules
Most molecules are fairly fragile. The loss of an electron usually means the destruction of the mol-
ecule. Once a molecule loses an electron it is either an ionized molecule (unbalanced charges),
another molecule (because it also loses a nucleus), or just individual atoms. Sometimes, if the molec-
ular bond is broken and leaves behind a different molecule, the new molecule can be studied (like
H2C6 breaking down into H2 and three molecules of C2). More often, though, the breaking of a
molecular bond is due to some strong radiative process that will likely break any newly formed
molecules and ionize any atoms.
There are many different molecules found in space. Most are carbon based, due to the fact that
carbon creates the most stable molecular bonds. It is thought that all life in the universe ought to be
carbon based because of this fact and the fact that there is so much carbon in the universe.
Interestingly, the molecules found in space are fairly complex. One might think that the radia-
tion from active stars and active galaxies and supernova remnants would make it difficult for
complex molecules to form, but astronomers have found sugar (poly-carbon structures, like gly-
colaldehyde, H2COHCHO) in our own Galaxy. In fact, since sugars are strings of carbons with
hydrogen atoms attached, one could say (and be completely correct) that the Milky Way contains
carbohydrates!
DUST IN GALAXIES
The second of the three main constituents of a galaxy is dust. Interstellar dust
ranges in size from large molecules to small dust particles (a few hundred
microns in size). Long molecular chains of hydrocarbons, called polycyclic
aromatic hydrocarbons (or PAHs), are the smallest particles of dust. Large
dust grains are typically a few hundred microns (micrometers) across (less
than 1 mm) and may have layers of water ice on them. It is thought that most
dust grains are either made of carbon or some kind of silicate.
It is difficult to study the composition of dust directly since the only
observable features are the blackbody spectrum of the solid particles and
112 • STARS AND GALAXIES
Kinds of Ice
In astronomy, it is often necessary to differentiate among different kinds of frozen gases and liquids.
So, when astronomers talk about normal ice, they will usually say “water ice.” This is because there are
many other kinds of ice in the universe that astronomers can study. Even in our solar system, there
are ices of methane, ethane, and ammonia in the atmospheres of the giant planets and in the com-
positions of some comets.
the reflected and scattered light due to the presence of the solid particles.
The blackbody spectrum gives astronomers a way to determine the tem-
perature of the dust. From these data, astronomers can hypothesize how
the grains get heated to such a temperature and then build models to help
narrow the possible chemical composition of the grains. The reflected and
scattered light due to dust give astronomers some clues about the location
of most dust as well as possible sizes and even chemical compositions of the
dust grains. Obviously, there must be a lot of hydrogen in the dust grains
(since hydrogen is the most abundant element in the universe), but some
other element is required, chemically speaking, to create solid grains.
Figure 5.1This diagram demonstrates the Bohr atomic model. In this model the nucleus of the hydro-
gen atom consists of one proton. Orbiting around the nucleus is one electron. The electron is pro-
posed to have several allowed orbits located at different distances from the nucleus. Each orbit is
proposed to be circular. This model is not physically accurate—the electron does not orbit the proton
in a circular orbit at a fixed distance from the proton. However, despite this inaccuracy, the model is
useful for visualizing the changes in energy levels that the electron does undergo when photons are
absorbed or emitted. [Jeff Dixon]
The evidence that supports these claims is complex and difficult to com-
municate without first giving a laundry list of all the features of the extinction
curve and newly observed emissions of the interstellar medium; however, it
is important to note here that the extinction curve for our Galaxy is not the
same as the extinction curve for other galaxies (however, it is surprisingly
similar to the extinction curves of the most actively starbursting galaxies).
In fact, it is well known that most other galaxies probably have different dust
constituents and size distributions. For example, two of the galaxies nearest
to ours (the Large and Small Magellanic Clouds) do not appear to have the
graphite “bump” feature, and probably, therefore, do not contain as much
dust that is carbon based.
STARS IN GALAXIES
The third and final constituent of a galaxy is the stars. Most galaxies con-
tain a huge number of stars (several billion), although some have as few as
What Makes a Galaxy? • 115
10 million. Even with these large numbers of stars, stars do not dominate
the masses of galaxies. Of the three constituents described in this volume,
gas (specifically, molecular gas) is the greatest contributor to the mass of a
galaxy. The populations of stars that make up a galaxy reflect the galaxy’s star
formation history.
Galactic Mass
In fact, none of the constituents of a galaxy mentioned so far dominate the mass of a galaxy. The
dominant contributor to the mass of a galaxy is something known as dark matter. The discovery of
dark matter, as well as current hypotheses about its nature will be discussed in detail in chapter 6.
Some galaxies contain both old and young stars. These galaxies have
been forming stars more or less continuously since the galaxy was formed.
Many galaxies like our own Galaxy are like this, containing both older and
younger populations (including the highest mass main sequence stars), as
well as active star-forming regions where objects that are not yet considered
stars exist.
Other galaxies contain only old stars (no main sequence stars with life-
times less than 1 billion years, for example). Such galaxies stopped forming
new stars some time ago (1 billion years ago, in the example above) so that
the only stars in the galaxy are stars that formed before the time when stars
ceased to be able to form. Of course, in these galaxies, there are no active
regions of current star formation. In fact, we also find that in these galaxies
the ingredients needed for star formation are not present in the necessary
proportions.
Still other galaxies contain only a few stars and are predominantly made
of gas and dust. These extremely faint galaxies are fairly new discoveries (in
the late 20th century), although some astronomers hypothesized that such
galaxies might exist decades before they were discovered. These low surface
brightness galaxies also appear to have the ingredients for star formation
unbalanced. Since these galaxies are such new discoveries, they have not
yet been studied well enough to determine whether these galaxies recently
formed stars or stopped forming stars a long time ago.
GALAXY SIZES
Galaxies, like stars, come in a range of sizes and shapes. In this section of
this chapter, we will explore the range of sizes that known galaxies exhibit.
There may be a wider range of galaxy sizes than what is discussed here, but
in the early 21st century, this encompasses what is known about the possible
range in size of galaxies.
116 • STARS AND GALAXIES
The smallest galaxies are barely larger than a cluster of stars. In fact, it may be
possible that some of the objects currently called dwarf spheroidal galaxies
may be runaway giant globular clusters. The smallest known galaxies were
discovered just a few years ago in a nearby cluster of galaxies, known as the
Fornax cluster. A team of Australian astronomers took images and spectra of
these newly named ultra compact dwarf galaxies.
Ultra compact dwarf galaxies contain a few tens of millions of stars and are
only a few tens of parsecs across. Yet, these tiny compact systems of stars con-
tain all the constituents of a galaxy. Their extremely small size makes them
about 10,000 times less luminous than the average galaxy.
Astronomers are still trying to understand whether these galaxies are the
building blocks of larger galaxies or whether they are the tiny cores of small
galaxies that have been stripped of their discs through some violent interac-
tion. Current theories of galaxy evolution suggest that there should be more
small galaxies than large ones, as our current understanding of the formation
of the universe and the evolution of galaxies involves building large galax-
ies from small pieces. Thus, there should be lots more of the small pieces of
galaxies still available to build larger galaxies. The recent discovery of ultra
compact dwarf galaxies may turn out to be new evidence to support the cur-
rent paradigm of galaxy evolution.
The average galaxy contains close to a billion stars, along with the normal
proportions of gas and dust in their many different forms. Usually the gas
and dust are located only in a disc, while the stars are spread more uniformly.
However, not all galaxies are disc dominated (as we will see in more detail
below), so sometimes the gas and dust are more spherically distributed. The
average galaxy has an absolute magnitude of -20. This means that the average
galaxy has a luminosity of about 1040 erg/s or 1034 Watts.
Magnitude Scales
The magnitude scale is a logarithmic scale of brightness. There are two parallel such scales in use—
absolute magnitude and apparent magnitude. Absolute magnitude is a measure of the brightness of an
object at a fixed distance from earth of 10 parsecs. Apparent magnitude is a measure of the brightness
of an object at its actual distance from earth. To measure absolute magnitude, one must know or cal-
culate or determine in some way the luminosity (energy per second) emitted at the surface of the star.
(For more information on these scales see chapter 2 of this volume.)
What Makes a Galaxy? • 117
Parsecs
A parsec is the distance to an object that has a parallax of one arcsecond. This distance is the equiva-
lent of about 3.3 light years, or the distance light can travel in 3.3 years, or about 3.6 x 1016 m.
(For more information on the parsec, see chapter 2 of this volume.)
The largest known galaxies are the giant elliptical galaxies. These galaxies
can contain trillions of stars. Giant elliptical galaxies are found in the centers
of dense galaxy clusters and are thought to have formed through many merg-
ers of smaller galaxies. Since giant elliptical galaxies have very little gas and
dust, their evolution must have been violent enough to trigger massive star
formation episodes that quickly exhausted all the fuel for making new stars.
Still, such galaxies show many different epochs of star formation (distinctly
different-aged populations of stars). This fact implies that such periods of
massive star formation must have recurred periodically. The model for giant
elliptical galaxy formation being due to many different mergers, therefore, is
supported by this evidence.
Giant elliptical galaxies are close to 10 times brighter than average galaxies
and can be up to 2 million parsecs across. This is larger than most groups of
galaxies and very close to the size of many clusters of galaxies. As you might
have guessed these galaxies are not irregular in shape at all, rather, they are
quite regular and do not include a disc component to their structure.
GALAXY MORPHOLOGY
In the beginning of the 20th century what is now known as a galaxy was
called a “nebula” (pronounced “neb-you-lah”). The word nebula means
“cloud.” In those days, astronomers thought that all the fuzzy objects they
were seeing were the same thing—nearby clouds of gas and dust with a few
stars illuminating them. Some of the things astronomers named nebulae (the
plural of nebula; pronounced “neb-you-lee”) turned out to be just what they
thought—clouds of dust and gas; however, many of these objects were gal-
axies. In fact, each object astronomers called a “spiral nebula” was, in fact,
118 • STARS AND GALAXIES
a galaxy. Today, these objects are known as spiral galaxies. Spiral galaxies are
just one of the many kinds of disk galaxies known.
Before astronomers even understood what galaxies were, the morphologi-
cal classification of these objects was already underway. The morphology of
a galaxy is a study of the galaxy’s shape. The root “morph-” means “shape”
in Greek. Since astronomers in the late 19th and early 20th centuries had
telescopes powerful enough to see many galaxies, they began to classify what
they saw by their shapes. Without understanding more than that there were
many of these objects and they appeared to have different shapes, astronomers
went about trying to understand them by separating them into groups.
This process is very similar to the process we all use to learn. First, we
see something we don’t understand, then we try to understand it by find-
ing similarities and differences between the thing we don’t understand and
things we do understand. Finally, we learn enough about the new thing to
be able to identify it as a thing different from everything else we know, but
similar to many different things (perhaps) that we do understand. This is the
beginning of the scientific way of knowing. Scientific knowledge requires
consensus, or agreement, among all observers. So, when an individual sees
something new, she should be able to describe it in terms of things she, and
others, know and understand. That way, others who find the same thing will
be able to identify it.
Galaxies are most often described by their shapes, with some emphasis on
any special relevant feature(s). The first classification scheme, developed by
Edwin Hubble (after whom the famous Hubble Space Telescope is named),
implied that there was an order within the classifications indicating an evolu-
tionary sequence. Hubble hypothesized that galaxies started out as ellipticals
and flattened over time becoming spirals in the end. It is still hypothesized
today that the different shaped galaxies are, in fact, fundamentally different
from one another; however, there is still little evidence to support this claim,
since galaxy evolution is not entirely pieced together enough to confirm this
hypothesis. For all we know, at this point in time, all galaxies started out the
same way, but over time, due to their environments, or due to some other
factors we don’t yet understand, they changed and evolved differently, ending
up with the various shapes and sizes we now observe.
All that aside, according to the most recent and best understanding of gal-
axies, they can be divided into three different groups. These are disk galax-
ies, spheroidal galaxies, and other galaxies. Originally, when galaxies were
classified, there were more than three morphological groups. There were
spirals, barred spirals, ellipticals, irregulars, peculiar galaxies, and more. In
this section of this chapter, these different types will be discussed, but rather
than presenting a zoo of galaxy shapes, the morphological types have been
grouped together to emphasize some of the newest advances in understand-
ing galaxy evolution that have occurred in the last few decades.
Almost a century after galaxies were first grouped together by shape,
astronomers have learned much more about galaxies and understand far
What Makes a Galaxy? • 119
more than just the shapes of galaxies. The content of galaxies varies with
morphology. Astronomers have inventoried the stars in several of the nearby
galaxies. They learned that the stellar content of galaxies varies with mor-
phology. The variation in stellar content tells astronomers something about
galaxy evolution when stellar evolution is taken into consideration. Gas and
dust masses vary with galaxy morphology. Additionally, star formation rates
can be estimated for many galaxies.
The dynamics of a galaxy varies with morphology. Astronomers have
measured the speeds of the gas and stars in many different galaxies of many
different types. It has been shown that disk galaxies (spirals and barred spi-
rals) have completely different kinematics (motions) than spheroidal galaxies
(ellipticals). All these things together demonstrate in a fairly clear way that
Hubble’s hypothesis cannot be correct. Galaxies do not start out spheroidal
in shape and gradually flatten over time. Still, many astronomers hypothesize
that there may be some kind of evolutionary progression somehow discern-
able through Hubble’s original morphological classification. Although, there
is no clear evidence to support this hypothesis, yet.
So much more is known today than 100 years ago that the old morpho-
logical classifications seem almost tedious to present. The fact that galaxies
can be grouped by shape and still be interesting to discuss is intriguing. It
certainly implies that the shape of a galaxy must be significant. It is quite pos-
sible that the century-old morphological classification scheme can yet teach
us something new.
DISK GALAXIES
Disk galaxies are those galaxies that have a component of their constituents
rotating in a two-dimensional disk. This type of galaxy is, perhaps, more
common than one might think. Originally, it was not known that the galaxies
classified as spiral and barred spiral were, in fact, disk galaxies. Astronomers
hypothesized that spiral and barred spiral galaxies must be flattened into a
disk-like shape, but this was largely based on observations of many galaxies
with spiral features that were viewed from different angles. The supposition
was that all galaxies with spiral features were pancake-like in shape, but seen
face-on, rather than edge-on. The counter supposition also applied: those
galaxies seen edge-on were supposed to have spiral features, although they
were not observable from the apparent angle.
In the last century, astronomers have been able to acquire data that sup-
port these hypotheses. So the idea of putting spiral galaxies into the same
category as flat galaxies was entirely justified, in the end. It is quite interesting
that very simple suppositions have turned out to be correct, when discuss-
ing the universe. It is probably for this reason that Occam’s Razor is such an
important principle for astronomers, in particular.
120 • STARS AND GALAXIES
Occam’s Razor
Occam’s Razor is the principle that states “entities should not be multiplied unnecessarily” (Merriam-
Webster OnLine). A 14th-century logician named William of Occam is attributed with devising this
principle. William was a Franciscan friar and used this principle to justify many different things,
including his belief that “God’s existence cannot be deduced by reason alone” Isaac Newton adopted
Occam’s Razor for his own purposes and rewrote it as “we are to admit no more causes of natural
things than such as are both true and sufficient to explain their appearances” (Physics and Relativity
FAQ, 1997).
A disk galaxy, in the most basic sense, has four main observable features:
a disk, a bulge, arms, and a halo. The disk is the pancake-shaped part of the
galaxy. It is thin and extends from the bulge outward. Within the disk is the
majority of the gas and dust contained in the galaxy. The bulge is a spheroi-
dal feature in the center of the disk galaxy. While there is some gas and dust
in most bulges, a bulge is predominantly stars. The arms are bright features
found within the disk usually beginning at the edge of the bulge and extend-
ing to the outer edge of the disk in a long arc. Finally, the halo is the spheroi-
dal component that encompasses all of the other parts of a galaxy. The halo
also contains gas, dust, and stars, but is predominantly comprised of dark
matter. (The nature of dark matter will be discussed in much greater detail
in chapter 6.)
Our own Milky Way galaxy is a disk galaxy. You can see this for yourself
by observing the Milky Way on a dark summer night. What we can see of our
galaxy from our vantage point is the disk. The disk of the Milky Way appears
as a thin band of stars and gas and dust across the sky. You can see more of
the Milky Way from the southern hemisphere. From there, the bulge of the
Milky Way is visible. The Milky Way’s bulge is small, indicating that it is a
truly disk-dominated object.
Normal Spirals
Normal spiral galaxies are disk galaxies with round bulges in their centers.
These bulges can vary in size from a very small fraction of the disk to so
much of the disk it is almost difficult to tell there is a disk at all. These gal-
axies are called spirals because the dominant features observed in the disks
of these galaxies are the so-called spiral arms. These are bright features that
extend from the bulge out into the disk in a long arc. Where the spiral arms
end is usually considered the end of the visible part of the disk.
The spiral arms are the most prominent feature of a normal spiral galaxy.
The nature of this feature will be discussed in much more detail in chapter 7
of this book. For now, we will say just that the arms consist of the same con-
Figure 5.2This diagram shows the parts of a spiral galaxy. The flat, round part of the gal-
axy is called the disk, the central, spheroidal part is called the bulge. The surrounding
material that is both above and below the disk in a spheroidal distribution is called the
halo. When observing the disks of some spiral galaxies face-on (so they appear round)
one can see the spiral arms. These are long bright parts that extend from the bulge to the
outer parts of the disk. [Jeff Dixon]
122 • STARS AND GALAXIES
stituents of the galaxy: gas, dust, and stars. There are no more stars in the
arms than in between the arms, but the types of stars in the arms are different
from the ones between the arms. The stars in the arms are the youngest, most
luminous stars in the galaxy. This explains, somewhat, why the arms are the
most prominent feature of a spiral galaxy.
There are some spiral galaxies whose arms are not very prominent. Yet,
the galaxies are still called “spiral.” This is because of the historical nature of
galaxy morphological classification. Originally, all disk galaxies were called
spiral galaxies. Later, astronomers learned more about the nature of disk gal-
axies and many new morphological classifications became apparent. Now, we
still call all disk galaxies spiral, but we have other adjectives we use to refine
that classification. However, a spiral galaxy in the most general terms really
means just that it’s a disk galaxy, since not all spiral galaxies have prominent
spiral arms.
Among the normal spiral galaxies, Hubble noted that the bulge size var-
ied from large to small. So, he devised a scheme to classify galaxies by arm
prominence, arm angle of separation, and bulge size. The symbol used for the
classification of spirals is S. The subclasses are indicated by letters a–d. So,
spiral galaxies can be Sa, Sb, Sc, or Sd, or anything in between (e.g., Sab, Sbc,
etc.). The “a” galaxies have the largest bulges, while the “d” galaxies have the
smallest bulges. Due to the attachment of Hubble’s scheme to his hypothesis
about galaxy evolution, Sa galaxies are often called “early type” spirals, and
Sd galaxies are often called “late type” spirals.
As mentioned before, our Milky Way galaxy is a disk galaxy. Although
it is not a normal spiral, the bulge size indicates that it is a “late type” spiral
with a subclass of cd. The “cd” subclass indicates that the bulge of the Milky
Way is not the smallest, but is a size in between the smallest and the next
largest size.
Barred Spirals
Barred spirals are the first well-documented new morphology for spiral gal-
axies. Early in the era of galaxy morphological classification, it was observed
that some spiral galaxies appeared to have elongated bulges. The center of
the galaxy was not round, but bar-like, extending out into the disk. Even the
arms seemed different, attached only to the ends of the central bar.
In the beginning, it was thought that there was some fundamental differ-
ence between so-called normal spirals with round bulges and the new barred
spirals with elongated bulges. Today, it is understood that the bar feature is a
transient feature in spiral galaxies. Astronomers are beginning to understand
many of the different ways bars can form and sustain themselves in disk gal-
axies. Each of these methods must be long-lived, as the vast majority (66%)
of all known disk galaxies are observed to have bars to some extent.
What Makes a Galaxy? • 123
Figure 5.3 This image of NGC 1300 shows what a typical barred galaxy looks like. Notice that the arms
of the spiral begin at each end of the bar, not on the sides or corners. [Credit: Hubble Heritage Team,
ESA, NASA]
Our Milky Way galaxy is a barred spiral galaxy. The central bulge of the
Milky Way is not round, but peanut shaped. The peanut shape is a common
bar shape for small or weak bars. The unusual shape of the Milky Way’s bulge
was discovered in the 1990s. Without very high resolution infrared imag-
ing (available only recently), it was difficult to demonstrate that the bar was
indeed a bar and not an oval or some other elongated shape. More recent
data from the Spitzer Space Telescope points to a very large and prominent
bar at the center of the Milky Way.
Giant Spirals
While most spirals are average in size, there are several well-known giant spi-
rals. The most well-known example of a giant spiral is the Milky Way. While
average-sized spirals might have a billion stars, giants, like the Milky Way,
have hundreds of billions of stars.
Usually giant galaxies have several satellite galaxies. Satellite galaxies are
not exclusively around giant galaxies, but there are very few giant galaxies
without known satellites. A satellite galaxy is usually a dwarf galaxy that is
in orbit around another galaxy. Satellite galaxies are gravitationally bound to
the giant galaxies they orbit.
124 • STARS AND GALAXIES
Figure 5.4This diagram shows what an elliptical galaxy looks like. This shape is not the
most flattened shape, but it shows a high aspect ratio so that one can see that one dimen-
sion is much larger than the other two. [Credit: NASA, ESA, and The Hubble Heritage
Team (STScI/AURA)]
SPHEROIDAL GALAXIES
The other dominant shape for galaxies in the universe is spheroidal. These
galaxies, notably, have no disks and have a shape that is three-dimensionally
round. Originally, these galaxies were called elliptical galaxies. This name
was given to these galaxies because scientists thought that they were see-
ing the same shape galaxy from different angles. So, when a galaxy appeared
circular, it was assumed that this galaxy was elongated in the dimension cor-
responding to depth.
Spheroidal galaxies, as a group, contain much less gas and dust than disk
galaxies. Some spheroidal galaxies appear to be almost completely devoid of
these constituents. Spheroidal galaxies are the largest and smallest galaxies in
the universe. Also, spheroidal galaxies are the most common type of galaxy
in the universe. This is the case mostly because the vast majority of dwarf
galaxies are spheroidals and there are far more dwarf galaxies than average-
sized or giant galaxies in the universe.
For a long time it was thought that dwarf spheroidal galaxies were the small-
est galaxies in the universe. These galaxies contain a few hundred million
stars and are usually three-dimensionally round, rather than cigar shaped,
for example. Dwarf spheroidal galaxies are usually a few thousand parsecs
in diameter. These galaxies are almost always found orbiting a larger galaxy.
There are only a handful of dwarf spheroidal galaxies that have been found
to be isolated.
In the beginning of the current century, a new, even smaller spheroidal
galaxy was discovered. This new galaxy contains only 10 million stars and
they are less than a hundred parsecs in diameter. The density of stars in these
galaxies is so high that they have been dubbed “ultra compact galaxies.”
What Makes a Galaxy? • 125
Elliptical Galaxies
Elliptical galaxies are so named because when it was discovered that nebulae
(nebulae are what galaxies were called before they were known as galaxies)
could be round, but not have spiral features, it was thought there could be
only two kinds of galaxies—flat ones with spiral features and ellipsoidal ones.
The three-dimensional ellipsoidal shape of elliptical galaxies is hypothesized
by astronomers. The reason the three-dimensional shape of elliptical is not
known is that all very distant objects appear two-dimensional. To observe the
depth dimension of an elliptical galaxy, we would have to be able to observe
these objects simultaneously from two separate points in space very far apart
(farther apart than the distance from Earth to the next nearest star: Proxima
Centauri).
Depth Perception
Depth perception is due to the ability to observe simultaneously from two different points in space.
The human being has two eyes separated by a small distance (usually about 8–10 cm). This distance
difference allows us to perceive depth. The trick is that the view from each eye must have a region of
overlap on the other eye’s view. If there is sufficient overlap, the brain must reconcile the two different
images it is receiving from each eye. This is how humans can tell that one thing is farther away than
another, without being able to detect the images overlapping. This principle is used in astronomy to
measure distance using the parallax method. This method requires that one observe the same star
when the Earth is at two different points in space during its revolution around the Sun. These obser-
vations are not simultaneous, but they allow astronomers to measure the distance to stars within
about 500 pc of Earth.
OTHER GALAXIES
Aside from these two major types of galaxies, those with disks and those
that are spheroidal in shape, there are other identifiable morphologies. These
other galaxies are not nearly as common as the disks and spheroidals, but
What Makes a Galaxy? • 127
they are common enough to mention and describe in some detail. For the
most part, these other galaxies are some combination of the disk and sphe-
roidal types; however, some of these other galaxies are fundamentally and
kinematically different from both disks and spheroidals.
Lenticular Galaxies
The first of these other types of galaxies to be discussed are the lenticular
galaxies. These are the “in between” galaxies. They appear to have a weak disk
component, yet meet almost all the other criteria to be classified as spheroidal
galaxies. When galaxies were first classified by their shapes by Edwin Hubble
in the early part of the 20th century, he proposed a possible evolutionary
sequence (much like the stellar evolutionary sequence). Hubble thought
(quite reasonably) that galaxies must start out very round (as spheroidals)
and flatten out over time (becoming disks). In Hubble’s scheme, lenticulars
were the “missing link” showing that the galaxies could be anywhere in the
spectrum of disk to spheroidal.
Hubble’s evolutionary sequence has not been shown to be correct or incor-
rect. The evolution of galaxies is still being pieced together. The timescales
for change in galaxies is much longer than that for stars. Understanding mor-
phological evolution, if it exists, will take good modeling and a much better
understanding of morphology and how it relates to the physical nature of the
galaxy than that which currently exists.
Irregular Galaxies
Irregular galaxies are nothing like either disk galaxies or spheroidal galaxies.
For the most part, these are dwarf galaxies. They look a little like the central
parts of disk galaxies (the bulges or bars). The most common example of an
irregular galaxy is the Large Magellanic Cloud. This is a dwarf galaxy that
orbits the Milky Way. It is one of the closest neighbors to our galaxy.
The Large Magellanic Cloud has a bar and a lot of gas and dust (much like
a disk galaxy), but no disk (like a spheroidal galaxy). There is a lot of recent
and ongoing star formation in this galaxy, and at the same time, there is an
underlying old stellar population in this galaxy, indicating that this galaxy
has been around for a long time. Kinematically, irregular galaxies tend to
look like disk galaxies, in that most of their constituents are moving in two
dimensions.
Peculiar Galaxies
This type of galaxy is by far the most unusual. Most galaxies classified as
peculiar are actually not really a single galaxy, but a pair (or more) of gal-
128 • STARS AND GALAXIES
RECOMMENDED READINGS
WEB SITES
Our own galaxy is called the Milky Way. As mentioned earlier, the name for
our galaxy in English is related to the Greek myth describing how it came to
be. The myth says that the Greek Goddess, Hera, Queen of the Gods, while
nursing Heracles spilt her breastmilk in a streak across the sky, creating the
Milky Way (“galaxias” in Greek).
In the previous chapter, we discussed the many different kinds of galaxies
that astronomers know about. What kind of galaxy is the Milky Way? How
do we know what kind of galaxy it is? In the previous chapter, we learned that
galaxies are so enormous that humans have yet to travel far enough to leave
the Milky Way.
In fact, if you imagine the solar system taking up a football field, our Gal-
axy would be 50 million times as large—approximately equivalent to 10 times
the distance between Earth and the Moon. Interestingly, the furthest humans
have traveled in space is to the Moon. On the football field solar system, that
distance is equivalent to one centimeter. Humans have hardly explored the
solar system, let alone the Milky Way galaxy!
In this chapter we will discuss what is known about the Milky Way in the
context of the other galaxies we can study. Even though the Milky Way is
our nearest example of a galaxy (just as the Sun is our nearest example of a
star), astronomers do not describe other galaxies in comparison to our own
as they talk about other stars in comparison to the Sun. Rather, we know less
of our galaxy because of our position within it. We cannot get a good view
of the whole. It is much like trying to determine the extent of a forest while
you are standing within it. So, studies of the Milky Way that have produced
interesting new discoveries have occurred, perhaps, more recently than one
might expect.
129
130 • STARS AND GALAXIES
In the late 18th century, a famous brother and sister team of astronomers
attempted to determine our position in the universe. William Herschel was
a musician who became interested in astronomy through books. He enlisted
his sister Caroline to assist him in his pursuit of knowledge through astron-
omy. He started by building telescopes. Shortly after he began observing
using the telescopes he built, he discovered the planet Uranus. As a result he
was employed as the Court Astronomer by King George III of Britain.
After making several discoveries of comets and natural satellites around
the giant planets, William and Caroline concentrated their efforts on attempt-
ing to determine the “Condition of the Milky Way.” At that time, the concept
of the Milky Way was entirely different from what we know today. It was
thought that what one could see in the night sky constituted the whole of
the universe. The other objects one could see using the most powerful tele-
scopes of the time were thought to be part of the Milky Way. That is, it was
not thought that other galaxies existed outside of the Milky Way. Although
by this time, astronomers had discovered many objects, now known to
be galaxies, at the time they were dubbed simply “nebulae.” Astronomers
assumed they were clouds of gas that were within the Milky Way, like the
Orion Nebula.
So the Herschels went about determining the “condition of the Milky Way”
by attempting to map the Milky Way. They made a simple assumption. They
assumed that they could see all the stars in the Milky Way. To determine the
extent of the Milky Way, one simply had to count the stars one could see.
Of course, we now know this is not the case, but in the late 18th century,
astronomy was still in its infancy. Photography had not yet been invented,
and an excellent astronomer was someone who could draw well what they
were seeing through a telescope or someone who could be very meticulous
about their measurements of time and position in the sky.
Given the information William and Caroline had about the universe, this
was a good assumption to make. They had no reason to believe that stars
could have different intrinsic brightnesses or that a star’s light might not
reach the Earth. So, in their crude attempt to determine the extent of the
Milky Way, they found that the galaxy appeared to be irregularly shaped, but
slightly elongated along one axis with the solar system located near, but not
exactly at the center. From their measurements, the Milky Way extended to
the location of what is now known to be the center of our Galaxy on one side
and to the outer edge on the other side.
What the Herschels did not take into account was the existence of inter-
stellar dust, which can obstruct the light from a star, hiding it from view to an
observer on Earth. Again, it is not that William and Caroline were not smart
enough to notice that the universe is dusty. Rather, it is that they had no way
of making accurate measurements of how much light was missing due to
intervening dust absorbing or scattering a star’s light.
The Milky Way • 131
Figure 6.1 This is the map of stars that the Herschels made. The large dot near the middle of the dia-
gram is the location of the Sun. According to what the Herschels were able to measure, it looked like
the Sun was located very close to the center of the Milky Way. Shortly after this map was constructed,
the Herschels discovered that interstellar dust affected what they were able to see. Once they had
accounted for the interstellar dust, their understanding of our position within the Milky Way was
modified. [Image courtesy History of Science Collections, University of Oklahoma Libraries.]
Caroline Herschel
Caroline Herschel was born in Hanover, Germany, in 1750 to Issak and Anna Herschel. Issak was a
musician (although he did other work to support his family) and hoped that all six of his children
would become musicians as well. Caroline was, however, to become her mother’s maid. At age 10 she
contracted typhus. As a result, her growth was stunted—her adult height was four feet three inches.
Her father believed that since Caroline was so small and homely, she would never marry.
At age 22, however, Caroline’s elder brother, William, rescued her from Hanover claiming her as his
own maid; instead, however, he gave her voice lessons. She became a prominent performer in Bath,
England, where she lived with her brother. When William decided to pursue his studies of astronomy
and telescope building, he trained his sister in mathematics and telescope building. At the age of 32,
Caroline was her brother’s apprentice. King George III even gave her a pension of £50 for her work in
astronomy.
William’s notoriety continued to grow, and Caroline began to do more and more on her own. She
detected several nebulae, catalogued them, and discovered eight comets. William married and came
132 • STARS AND GALAXIES
to the observatory he had built less and less frequently over time. When William died, Caroline
ended her career as an observational astronomer and returned to Hanover, Germany, to live out the
rest of her life. Before she died, she catalogued every discovery William and she had made and was
proclaimed an honorary member of the Royal Astronomical Society. She was the first woman to ever
receive this honor. Caroline Herschel died in 1848 at the age of 98.
Figure 6.2 This diagram shows the location of the globular clusters in the halo of the Milky
Way, far from the interstellar dust located in the disk. Using globular clusters to map out
the distribution of the stars in the galaxy avoids the problem of the interstellar dust. It is
much like using a light house to navigate around rocks on a foggy ocean. [Credit: NASA,
ESA, and The Hubble Heritage Team (STScI/AURA).]
So, Shapley observed many globular clusters and concluded that the loca-
tion of Earth is not in the center of the galaxy. But, this time, unlike in Coper-
nicus’s and Galileo’s day, the general public was ready to accept the conclu-
sion that the Earth was not at the center of the galaxy.
THE DISK
The most obvious evidence that we live in a disk galaxy is the view we have of
our Galaxy. If you observe the Milky Way (best seen in northern hemisphere
summer), you will note that from Earth, what we see is a thin band of stars.
If we lived in a spheroidal galaxy, it would be impossible to view our Galaxy
from an angle that would make it appear as a thin band of stars. Rather, a
spheroidal galaxy from the inside would look like a roundish distribution of
stars in the night sky. Depending on where were located within a spheroidal
galaxy we might see about the same number of stars in every direction, or we
might see more stars in one part of the sky (where the center of the Galaxy
would be located) and a relatively even distribution of stars in almost every
other direction.
To determine how it is that we see a band of stars from within a disk gal-
axy, picture a CD and imagine yourself somewhere within the disk. The disk
of a disk galaxy is thick enough that many stars (perhaps up to 1,000) could
The Milky Way • 135
exist in the same part of the disk. A good visual analogy that gives an exag-
gerated thickness to the disk is to imagine that you are a blueberry inside a
giant blueberry pancake. Since the blueberry is part of the pancake, the view
of the pancake from the perspective of the blueberry would be just the inside
of the pancake nearest itself. However, looking carefully, you would notice
that there is little or no pancake above or below the blueberry. Since our view
of the Milky Way is a thick band of stars (just like the view of the pancake to
the blueberry), we must be inside the disk of our Galaxy.
The disk of our Galaxy is where the majority of the bright astronomical
objects are located. The brightest stars are in the disk of the Galaxy, bright
clouds of gas illuminated by bright stars are located in the disk of the Galaxy,
and bright dust clouds lit up by stars are located in the disk of the galaxy.
Most of what makes up a galaxy is located in the disk of the galaxy.
Open Clusters
Open clusters of stars are groups of stars that are physically related. In an
open cluster, all the stars form from the same diffuse nebula of gas and dust.
Nearly all the stars form at the same time. So, open clusters contain stars
with the same chemical composition that are the same age. The name “open
cluster” is given to these stellar associations because they are relatively diffuse
objects with no obvious center.
Open clusters are found in the disk of a spiral galaxy. Because of their
gravitational interactions with clouds in the disk of a spiral galaxy, open clus-
ters tend to diffuse (lose their members) in a short period of time, compared
to the lifetimes of the members. Most open clusters spread their members
along their path within a few hundred million years. Open clusters are con-
sidered young because the stars of which they are comprised are recently
formed. Also many of these kinds of star clusters still contain the dust and
gas that was used to form the stars within them; however, it is important to
note that the word “young,” here, does not mean a few years old; it means,
typically, less than 10 billion years old.
Open clusters are characteristically irregular in shape and contain usually
only a few tens to several thousand stars. Since open clusters are so diffuse,
they are not easy to spot. It usually takes a lot of stellar spectra to determine
the members of an open cluster, and even then, there are often members
left out or stragglers included accidentally. There are several hundred known
open clusters in the Milky Way.
The importance of open clusters in astronomy is discussed in great detail
in chapter 3. To summarize, open clusters, along with globular clusters were
used to piece together stellar evolution. This was done by putting the members
of each cluster (whether open or globular) on a cluster H-R diagram. (H-R
Diagrams are discussed in great detail in chapters 2 and 3 of this volume.)
Because of the fact that the members of a cluster are the same age (since all
136 • STARS AND GALAXIES
THE ARMS
The arms of the Milky Way are where there are dense concentrations of gas,
dust, young stars, and star-forming regions. Primarily, the arms of a disk
galaxy are bright spiral-like features seen within the disk of the galaxy. The
arms are bright because the arms are the place where all the ingredients for
star formation are present in exactly the right amounts. Also, in the arms,
clouds of dust and gas are being pushed together, which causes the gas clouds
to break up into small pieces, which collapse further, forming stars. In chap-
ter 7, more will be said about how arms work and what their significance
is to the galaxy, dynamically speaking.
Star-Forming Regions
Star-forming regions are places were stars are forming. They are usually
clouds of hydrogen gas that are lit up by the young stars forming inside.
Most often, these regions are called HII (pronounced “H-two”) regions. This
is because HII is the astronomical symbol and term for ionized hydrogen.
In these star-forming regions the hydrogen is ionized by the intense ultra-
violet radiation coming from the most massive stars formed. The radiation
from these stars ionizes the gas surrounding these stars out to hundreds,
sometimes thousands of parsecs away from the stars themselves. So, what
astronomers see most often is a large cloud of ionized gas. Looking more
carefully, they may be able to see the actual stars responsible for ionizing the
gas they see.
These star-forming regions are important to study. What’s happening
inside them is a process about which astronomers still speculate. There are,
of course, physical models of the process, many of which rely on principles
of physics to guide them through the process, but in actuality, astronomers
still know very little about the conditions inside these star-forming regions.
Even today, simplifications are made to allow theoretical models to be
developed. The nature of these regions is still too complicated to unravel.
Astronomers are hard at work trying to unravel the secrets of star for-
mation, nonetheless. It is an important process to understand. Whether star
formation has temperature, density, or constituent limitations is still poorly
understood. Part of the problem is time. It takes a long time (several million
years) for star-forming regions to go from a cloud containing cold, neutral
gas to a cloud containing hot, ionized gas and still more time for the cloud
to become an open cluster or globular cluster. So, astronomers have to make
The Milky Way • 137
lots of assumptions about how these processes occur, since there are no data
about these regions from millions of years ago and an astronomer can not
wait millions of years to answer the question of how stars form.
Orion Nebula
The most well-known star-forming region is the Orion Nebula. This nearby, bright star-forming
region can be seen with a small telescope. Located in the “dagger” of the constellation Orion, the
nebula appears grey-green to the naked eye. However, many beautiful images of the famous nebula
can be found online. The Hubble Space Telescope has amassed an amazing image library containing
color-enhanced space-based images of the Orion Nebula. The ionized cloud of gas that makes up the
Orion Nebula is powered by four extremely bright stars.
The age of the stars in the Orion Nebula is approximately 1–2 million years, making it one of the
youngest clusters known in our own Milky Way. The famous nebula is also one of the closest star-
forming regions to Earth at a distance of about 500 pc (or 1,500 light-years). As it is one of the closest
star-forming regions, it should not be surprising that the Orion Nebula is also one of the smallest
known star-forming regions. Since most of the star-forming regions astronomers can study are very
far away, they are also very large. If those star-forming regions were not large, astronomers on Earth
would not be able to see them and study them.
Emission Nebulae
Star-forming regions are just one kind of emission nebula. Emission nebulae
are clouds of gas that emit light. Usually the clouds of gas are ionized and,
therefore, emit radiation due to the recombination of electrons with ions.
So, an emission nebula is found anywhere that gas is heated enough to be
ionized. Thus, emission nebulae are usually found near energy sources. Most
times, the energy source is visible (like a star), but sometimes the energy
source is not visible (like a supernova remnant or a black hole).
Star-forming regions are where many stars are forming all at once. In such
regions there are usually several massive stars forming. These stars will both
produce a lot of light and produce strong “winds” because of the extremely
high rates of fusion in their cores. This high rate of fusion can cause insta-
bilities in the star as it evolves. Sometimes, during the protostar phase, these
massive stars lose material because the radiation pressure is greater than the
pressure of gravity holding the star together, and the star blows apart.
Even in star-forming regions where such massive stars are not forming,
the presence of stars amidst gas alone is enough to create an emission neb-
ula. The light from the stars comes in all wavelengths. It takes photons of
wavelength 121.2 nm to ionize hydrogen gas (the most common gas in the
universe). All stars will radiate these photons since all stars are blackbodies.
Stars with higher surface temperatures will radiate more of these photons
138 • STARS AND GALAXIES
than stars of lower surface temperatures. So, the extent to which the gas
around a star is ionized will depend on the surface temperature of the star.
Assuming the gas around a star extends forever, a hotter star will ionize gas
to a greater distance than a cooler star.
Another type of emission nebula is a planetary nebula. A planetary nebula
is one of the end products of an intermediate-mass star (like our Sun). When
this type of star exhausts all the possible methods for fusion in its core, the
outer shells of the giant star expand during the last shell fusion phases and
never contract back to interact with the core. The shells that expand into
space are simply layers of (relatively) dense, hot gas that is ionized and emit-
ting light. The motion of this hot gas can also cause emission from gases with
which it interacts as it expands.
Additionally, a supernova remnant is a kind of emission nebula. A super-
nova is the event that marks the end of the existence of a massive star (one
with a main sequence mass greater than about eight times the mass of our
Sun). This event is caused by the outer shells of a massive star collapsing
back on its iron core. The core collapses further into a neutron star or a black
hole, depending on its mass. The shells bounce off of the collapsed core and
move outward into space at supersonic speeds. The shells are already hot,
ionized gas, and as they move with supersonic speed through the inter-
stellar medium, they collide with neutral gas and ionize it. This collisional
excitation (when atoms are excited or ionized because they are involved in
a collision with another particle) is distinguishable from photoionization
(when an atom is excited or ionized because it interacts with an energetic
photon). Collisional excitation causes more low-energy transitions than
high-energy transitions, whereas photoionization does the opposite—it
produces more high-energy transitions than low-energy transitions.
Still another kind of emission nebula can occur around supermassive black
holes. Around these objects, which are found in the cores of some galaxies,
material is being accreted onto the black hole by means of an accretion disk.
The accretion disk is a thin disk of material that is slowly falling into the black
hole. The material in the disk is very dense and very hot. The heat comes from
friction and shearing forces acting on the material in the disk. Again, the cause
for the emission is not photoionization, but collisional excitation. The atoms
are colliding with one another and causing the electrons to either move up to
higher energy levels or be removed from the nucleus, ionizing the atom.
Finally, the last type of emission nebula is the neutral hydrogen cloud.
As discussed earlier, the neutral hydrogen clouds emits photons because the
hydrogen atom has two ground states. One ground state (the more common
one) occurs when both the electron and the proton are oriented so that their
spins are considered “up.” The other ground state occurs when the electron
and the proton have opposite spins. Since the proton is a particle that does
not easily change its spin, it is the electron that does the flipping. This tran-
sition from the common ground state to the ground state of lower energy
is commonly known as the spin-flip transition. The photon emitted has a
The Milky Way • 139
Dark Nebulae
A dark nebula is one that appears dark against the background. The back-
ground could be an emission nebula or other stars. These nebulae appear
dark, because they are dominated by dust which blocks out the visible light
coming from behind. The most famous dark nebula in the northern hemi-
sphere is the Horsehead Nebula, which is located in the Orion Nebula (a
well-known emission nebula located in the dagger part of the constellation
Orion). The Horsehead Nebula is difficult to see, even with the aid of a tele-
scope because it is very faint. It is best seen photographed for a long exposure
time, so that many photons can be gathered before the image is processed.
Our eyes don’t work that way—they integrate over a fraction of a second, so
they would need a very large telescope to gather more photons in our short
integration period to see the nebula. Such large telescopes don’t usually have
eyepieces for humans to look through—they are filled up with instrumen-
tation for astronomical studies. Against the faint emission nebula, there is
a dark feature that humans recognize as having the shape of a horse’s head. The
dark feature is really a dense cloud of dust.
In the southern hemisphere the most well-known dark nebula is the “Coal
Bin.” This is an extremely dark region of the sky. It is noticeably darker than
any other region not least because it occurs along the line of sight of the Milky
Way (which is spectacular in the southern hemisphere) so that it blocks out
the light of billions of stars. In comparison to parts of the sky that are con-
sidered dark because there are few nearby stars visible, the coal bin is darker
than these as well. This dark nebula is also a dense cloud of dust that is block-
ing the light from the stars behind it.
Reflection Nebulae
Reflection nebulae are found around young stars. Some of the material from
which they formed still surrounds the young stars and the dust that is behind
the stars, from our line of sight, is reflecting the light from the stars back in
our direction. The most famous reflection nebula is the one around the Ple-
iades. This reflection nebula makes the Pleiades look bluer than they really
are. This is because of the properties of the dust. The dust actually scatters the
light in such a way that the longer wavelength light continues through the
dust, while the shorter wavelength light is scattered so much that it goes back
in the direction from which it came. Since blue light is shorter wavelength
than red light, the nebula looks blue.
140 • STARS AND GALAXIES
Figure 6.3 This diagram shows how a reflection nebula and a dark nebula are just a dust cloud and a
light source (stars) seen from different points of view. If the dust cloud is behind the light source,
relative to the observer, the observer sees a reflection nebula where the blue star light is reflected back
at the observer. If the dust cloud is in front of the light source relative to the observer, the observer
sees a dark nebula where the light from the star is blocked by the dust cloud. [Credit; Diagram: Jeff
Dixon; Horsehead Nebula: JPL/NASA, NOAO, ESA and The Hubble Heritage Team, (STScI/AURA);
Pleiades: NASA, ESA and AURA/Caltech]
Light Scattering
When light interacts with dust particles in space, it is scattered. Light scattering is the cause of the
appearances of many astronomical objects. Dust scatters light so that the longer wavelength photons
are scattered the least and continue in the direction in which they were going with little change.
Conversely, shorter wavelength photons are scattered the most, so that they go back in the direction
from which they came.
This scattering of light is what makes stars seen through dust clouds are not visible or appear
much redder than they really are. By the same token, stars in front of dust clouds appear much bluer
than they really are because all the blue light they are emitting is being reflected back in our direction
by the cloud behind them, causing a blue haze to appear around the stars.
Interestingly, dark nebulae and reflection nebulae are made of the same
thing: dust. The dark nebula is how the dust appears when it is in front of the
light source, relative to the observer. The reflection nebula is how the dust
appears when it is behind the light source, relative to the observer.
THE BULGE
The bulge of the galaxy is what the central part of a disk galaxy is called. In
the case of the Milky Way, the shape of the bulge is difficult to determine,
since it is not clearly visible from our position within the disk of the Milky
The Milky Way • 141
Way. Typically, a bulge contains stars that move triaxially (with orbits cross-
ing through all three planes defining the spatial dimensions).
Towards the end of the 20th century, with the birth and growth of infrared
and submillimeter astronomy, it was beginning to seem as though the bulge
of the Milky Way might actually be a bar. About two-thirds of the bright
spiral galaxies in the universe appear to have a bar feature. This means that
most spiral galaxies have a bulge that is elongated, so that it looks like a bar,
rather than a bulge. The first evidence for a bar came in the 1980s and 1990s
with space-based and ground-based near and far-infrared observations of
the central regions of the Milky Way.
It is significant that the evidence came via the infrared regime of the electro-
magnetic spectrum. The primary reason that infrared detectors were needed
is that longer wavelength light can travel greater distances without being dis-
turbed (absorbed or scattered) by intervening media. That is, astronomers
see further through the pancake-like disk of the Milky Way when they collect
infrared light, rather than visible-wavelength light.
Second, it is important to understand that what astronomers are seeing
are very old stars that give off most of their light in the red part of the visible
spectrum or the infrared part of the electromagnetic spectrum. So, even if
there weren’t dust in the way to block the visible light, astronomers might
not have been able to see the stars in the center of the galaxy using visible
light detectors, because those stars are too faint in the visible part of the
spectrum.
The stars in the bulge/bar of the Milky Way are old. Most of the stars
in this part of the galaxy are so old that they are no longer main sequence
stars, like our Sun. They have evolved off the main sequence and are now
red giants. (For more on stellar evolution, see chapter 3.) The most recent
evidence has come from the Spitzer Space Telescope (SST). This is an infra-
red telescope that was launched in 2003 and has been orbiting the Sun, just
behind Earth in its own orbit of the Sun. From this vantage point, the SST
can see the universe and radio the data it takes back to Earth. Being further
from Earth means the SST doesn’t need as much coolant as it would if it
were in an orbit around Earth. This gives the telescope’s detectors a lon-
ger lifetime for less weight (a very important consideration for space-based
telescopes).
In addition to the many, many stars that make up the bulge/bar of the
Milky Way, there is, likely, a black hole in the center of our galaxy. The
evidence for the black hole is in the motions of the stars in the center of
our galaxy. Stars near the center of the galaxy (in the center of the bulge/
bar) are moving far faster than possible to explain using ordinary physics,
unless there is a massive object that is not seen in any part of the electro-
magnetic spectrum. Such an object would have a mass equal to several
million times the mass of our Sun and would be much smaller than our
solar system. The only object that could meet all three of these criteria is a
black hole.
142 • STARS AND GALAXIES
Globular Clusters
Globular clusters are very important indicators of the unseen part of the
Milky Way—the halo. Located above and below (as well as in) the disk of
the Milky Way, globular clusters are clusters of tens of thousands to mil-
lions of stars. The stars are gravitationally bound to one another and form
a roundish cluster that looks much like a tiny spheroidal galaxy. Like open
clusters, globular clusters probably formed in the disk, where the ingredi-
ents for star formation are plentiful. However, unlike open cluster formation,
globular cluster formation does not occur anymore. As far as astronomers
can determine, no globular clusters have formed in the last few billion years.
The youngest known globular cluster is thought to be 5–10 billion years
old, but its origin is suspect. It is entirely possible that this globular cluster
was not formed in the Milky Way, but was captured by the Milky Way more
recently.
The most obvious difference between open and globular clusters is their
size. Globular clusters are far more compact than open clusters, but globular
clusters contain far more stars than open clusters do. Secondly, their motions
are different. The open clusters remain within the disk, moving as other stars
in the disk move, whereas the globular clusters inhabit the halo part of the
galaxy. Finally, the stars in open clusters contain more metals (heavy ele-
ments) than the stars in globular clusters. This difference is thought to be due
to the fact that globular clusters formed earlier than open clusters, before
the gas from which stars are formed was enriched with metals by the cata-
strophic ends of the previous generation of stars.
Globular clusters are used in concert with open clusters to piece together
stellar evolution. Because stars change so slowly, it is not possible for a human
being to observe these changes in a lifetime (or even several lifetimes, for that
matter). So, to learn how stars evolve, it was necessary to use clusters. Because
all the stars in a cluster (either globular or open) form at approximately the
same time and because the stars can be considered to be all at the same dis-
tance from Earth, clusters of stars have been very helpful in unraveling this
mystery. For more information on the use of clusters in understanding stellar
evolution, refer to chapter 3.
RECOMMENDED READINGS
Freedman, Roger, and William J. Kaufmann III. Universe. 8th ed. New York: W.H.
Freeman Company, 2008.
Hawking, Stephen. On the Shoulders of Giants. Philadelphia: Running Press, 2002.
Seeds, Michael A. Astronomy: The Solar System and Beyond. 5th ed. Pacific Grove,
CA: Brooks Cole, 2006.
7
Arms in Disk Galaxies
The most prominent feature of a disk galaxy is its spiral arms. Sometimes
these are clearly defined arms, and sometimes the arms are just small
spurs appearing as though the disk were made of feathers. The arms are,
regardless of their appearance, where star formation is occurring within
the disk galaxy. Therefore, the arms are the most interesting part of a disk
galaxy.
The appearance of the arms in a disk galaxy vary considerably. Even if the
arms are prominent, the arms of a disk galaxy can have different sizes,
winding tightnesses, brightnesses, or lengths from galaxy to galaxy. Sym-
metry plays a big role in how the arms in a disk galaxy appear.
A galaxy with prominent arms that are clearly attached to the central
bulge or bar and can be followed continuously, spiraling outward until
they reach the edge of the visible disk is called a grand design galaxy. A
galaxy with arms that are unstructured and broken into parts so that the
disk appears like the fleece of a sheep is called a flocculent galaxy. There
are actually 12 different arm classifications varying between flocculent and
grand design; these classifications include possible causes for the differ-
ing appearances of the arms like the presence of a central bar or nearby
galaxies.
145
146 • STARS AND GALAXIES
Figure 7.1 The photograph on the left shows a flocculent galaxy. Notice that the arms are
not continuous; rather, they are in pieces or “spurs,” giving the disk a “fleecy” look. The
photograph on the right shows a grand design galaxy. Notice that the arms are long and
continuous from the central part of the galaxy to the outer edge. [Credit: NASA, ESA, S.
Beckwith (STScI), and The Hubble Heritage Team (STScI/AURA)]
massive stars will form in spiral arms, too. As a result, there is a higher den-
sity of massive stars on the leading edge of the spiral arms in a spiral galaxy.
As for the other components of the galaxy—gas, dust, lower mass stars—
these are distributed almost evenly throughout the disk of a spiral galaxy.
Even though these materials are in slightly higher abundance within the spi-
ral arm features of a spiral galaxy, there are gas, dust, and stars everywhere
within the disk of a spiral galaxy. So, the difference between the arm and
interarm environments is solely the density of the stars gas and dust that
make up the disk of the spiral galaxy.
ARM CLASSES
As mentioned earlier, there are 12 different arm classes that describe the
structure of the spiral arms in a spiral galaxy. Most of the arm classes
describe the presence (or lack thereof) of spiral arms, the number of arms,
the extent of the arms (from center to outer visible edge), and the appear-
ance of the arms, but some include information about nearby galaxies and
the presence of a central bar, rather than a bulge. Arm classes 1–4 are gal-
axies that are classified as flocculent galaxies due to their fleecy, disorga-
nized appearance. Arm classes 5–12 are galaxies that are classified as grand
design galaxies due to their long, prominent, symmetric arm structure.
The appearance of these galaxies tells the story of the underlying dynam-
ics. While most things in astronomy are not as they first seem, in this case,
the shape, length, prominence, and symmetry of the arms of spiral galaxies
reveal something about the physical conditions within that galaxy’s disk.
DENSITY WAVES
As discussed at length above, the spiral arms of a galaxy are not physical enti-
ties, but rather define a place within the disk where there is an over-density
of material. The material moves through this place creating new stars, which
makes this place seem as if there are more stars there than anywhere else in the
galaxy; however, there are just more new stars there than anywhere else in the
galaxy. Interestingly, though, the place is not stationery, either. The material
moves through the over-dense region and the over-dense region moves, too,
though usually much more slowly than the material moving through the region.
This movement of a region of over-density is described as a density wave.
The density wave theory is the explanation for spiral arms discussed in this
chapter. It further includes the fact that these waves are a result of gravita-
tional interactions within the disk combined with the underlying gravitational
potential well of the galaxy. The theory behind where the waves come from
is more tenuous. There is some evidence that density waves can be caused
by resonances in the orbits of stars in the disk of the galaxy, and there is also
148 • STARS AND GALAXIES
The winding dilemma is the famous explanation for why density waves must
be the correct mechanism for spiral arms. Because a large fraction of disk gal-
axies have spiral arms, we can conclude that spiral arms are long-lived features.
If they were not long-lived, we would expect that finding a galaxy with promi-
nent spiral arms would be difficult to do (i.e., spiral arms would be rare). Since
they are common, we conclude that they must be around for a long time.
So, the idea that these features might be physical becomes a problem. We
know, from observing the spectra of gas in galaxies (and even some stars in
galaxies), that the gas and stars move around the galaxy at a nearly constant
speed. In other words, stars near the center are moving with the same speed
as stars near the outer edge. If spiral arms were a solid, physical part of the
galaxy containing the same stars all the time, the spiral arms would wind up.
In the figure below, there are two examples of how material in a galaxy
could move. The example on the left shows solid-body rotation. In this case,
like a merry-go-round or Frisbee, the material near the center of the galaxy
moves so that the material near the outer edge is always on the same line.
Figure 7.2The two drawings demonstrate why density waves are necessary to explain spiral arm pat-
terns in galaxy disks. They also show the winding dilemma. Since the stars and gas in galaxies rotate
around the galaxy at the same speed, no matter how far they are from the center, this means stars that
were once in the arms would eventually move out of them. [Jeff Dixon]
Arms in Disk Galaxies • 149
This means that in equal intervals of time, the material on the outer edge will
go farther than the material near the center (like the second hand on an ana-
log watch). If galaxies rotated this way, the arms could be explained as solid
or physical features containing the same material all the time. However, we
know that galaxies don’t move that way.
Instead, galaxies move like the example on the right (called differential
rotation), so that in equal intervals of time, material near the outer edge of
the galaxy moves the same distance as material near the center of the galaxy.
This means that in less than one rotation of the material near the center of the
galaxy, the material on the outer edge has lagged behind. Take this forward
a few more rotations and your spiral arms will wind up, wrapping so tightly
that they can no longer be identified as spiral arms.
The rotation period of a galaxy is only a few hundred thousand years—less
than the length of existence of most stars. If the above scenario were true,
then on very short timescales spiral arms would disappear. Yet, we see spi-
ral arm patterns in many disk galaxies today. This is the winding dilemma.
Using this model of differential rotation, which is based on observed gal-
axy motions, it is clear that spiral arms cannot be explained as a solid entity
because the spiral features would disappear very quickly (within a few rota-
tions) and we would not expect there to be any spiral arm features visible in
the universe today (or at least, it would be very rare).
The theory of spiral density waves must account for both flocculent and
grand design galaxies. For that reason, simply constructing a model for how
they work in spiral galaxies is not sufficient—all mechanisms that cause spi-
ral arms must be nonfunctional in flocculent galaxies to pass muster. Fur-
ther, the surviving theory must explain all the varieties of spiral arms found
(arm classes 5–12), which run the gamut from a single arm to more than two
arms and from broad, fluffy arms to narrow, crisp-looking arms.
The current state of the spiral density wave theory is that a combination
of mechanisms probably work together to create the variety we see in spiral
structures. The grand design galaxies are most clearly explained using den-
sity waves without the need for an instability to drive the density waves. More
intermediate structures require a perturbation or instability within the disk
to explain the observed structure. Flocculents seem to be best explained by
stochastic star formation theories (not using spiral density waves at all).
structures. Instead of clear spiral arms, flocculent galaxies have loose, fleecy-
looking spurs of star formation that are disconnected from one another.
Stochastic star formation is the only proposed mechanism that successfully
explains and predicts flocculent galaxies.
In stochastic star formation, each event of star formation is involved in a
domino effect process such that it contributes to the next event of star for-
mation. As far as astronomers can determine, star formation is triggered by
some gravitational instability in a cloud. A gravitational instability can be
caused by many things. The process of stochastic star formation identifies the
cause of such instabilities as supernova explosions and stellar winds (both
from young, massive stars and from T Tauri stars).
During the process of star formation massive stars go through phases that
cause them to have extremely strong stellar winds. This is part of the massive
star’s journey to hydrostatic equilibrium (when the star is balanced by the
forces pushing it inward and those pushing it outward). During the mas-
sive star’s journey, the outward forces will sometimes overpower the inward
forces, causing massive ejections of stellar material at extremely high speeds.
These mass ejections can trigger star formation if they reach a cloud contain-
ing the right materials (enough molecular and neutral hydrogen) and cause
a gravitational instability.
Less massive stars, like the Sun, also undergo periods of instability that
cause strong stellar winds. This occurs during the T Tauri phase of stellar
evolution. During the T Tauri phase, stars eject material. This occurs because
they are accreting material from a protoplanetary disk and some of the mate-
rial being accreted is ejected in the form of jets (usually perpendicular to the
accretion disk). For more information about these stars, see chapter 4. The
jets from T Tauri stars can cause star formation to occur if their jets reach a
cloud containing the right ingredients and cause a gravitational instability.
In addition to these mechanisms, supernova events are massive explosions
of high mass stars. These explosions cause material to move extremely fast in
all directions from the point of origin. Since massive stars can only be formed
in large star-forming complexes where there is enough material to produce
such stars, supernova events will occur in those same locations, for the most
part. (Some exceptions include binary systems where mass transfer can occur
as well as Type I supernovae.) Supernova explosions were long thought to be
the only possible mechanisms for starting star formation. These explosions
move material at very high speeds. Such fast-moving material can start star
formation if it interacts with a cloud containing the right materials and cause
a gravitational instability.
In the above discussion about the winding dilemma, it was briefly mentioned
that galaxies rotate differently than a solid disk (like a merry-go-round
Arms in Disk Galaxies • 151
Doppler Effect
The Doppler Effect is an effect on waves due to motion. The most commonly known example of the
Doppler Effect is the sound a train makes as it passes you. (Actually, it works with any sound
source that is moving relative to you—an airplane, a race car, an emergency vehicle, a car with a
loud radio, etc.) In this example, as the source approaches the listener (or the listener approaches
the source), the sound waves are received in a compressed form. That is, the listener hears a
higher pitch. As the source recedes from the listener (or as the listener recedes from the source),
the sound waves received are in an extended form (stretched out). That is, the listener hears a
lower pitch.
Interestingly, Doppler presented this idea in the context of both light and sound waves as a pro-
posed explanation for why stars in binary systems have different colors. He proposed that their rela-
tive speeds would cause a shift in the light we received making them change color with their motion.
He was incorrect that this change in color would be great enough to perceive, but he was correct in
that this change in color does occur.
c λo-λe
v=
λe
where v is the speed of the source relative to the observer, c is the speed of
light, λe is the wavelength of the light emitted by the source, and λo is the
wavelength of the light observed by the observer. If the wavelength observed
Arms in Disk Galaxies • 153
is longer than the wavelength emitted, the light is being shifted to longer
wavelengths (towards the red end of the visible part of the electromag-
netic spectrum), meaning the distance between the source and observer is
increasing (either the source is moving away or the observer is). If the wave-
length observed is shorter than the wavelength emitted, the light is being
shifted to a shorter wavelength (towards the blue end of the visible part of
the electromagnetic spectrum), meaning the distance between the source
and the observer is decreasing (either the source is moving towards the
observer or the observer is moving towards the source).
Figure 7.3The graph on the left shows a mass distribution curve for the solar system. Notice that the
mass of the material within a certain radius decreases exponentially as the distance from the Sun
increases. This means that as you go farther out from the center, the amount of mass decreases. The
graph on the right shows a rotation curve for the solar system. Notice that the material closer to the
center moves more quickly than the material farther from the center. Again, the decrease in rotation
rate is exponential with increasing radius, mimicking the mass distribution. [Jeff Dixon]
154 • STARS AND GALAXIES
Figure 7.4 The graph on the left shows a mass distribution curve for the solar system. Notice that the
mass of the material within a certain radius decreases exponentially as the distance from the Sun
increases. This means that as you go farther out from the center, the amount of mass decreases. The
graph on the right shows a rotation curve for the solar system. Notice that the material closer to the
center moves more quickly than the material farther from the center. Again, the decrease in rotation
rate is exponential with increasing radius, mimicking the mass distribution. [Jeff Dixon]
mass distribution in such an object is fairly uniform. That is, there is the same
amount of mass (approximately) in 1 cm2 at the center of a Frisbee as there is
in 1 cm2 at the edge of a Frisbee.
The solar system moves with Keplerian motion. This means that the
bodies in the solar system follow Kepler’s laws of planetary motion, which,
loosely, state that objects that are farther from the center (Sun) move slower
and objects that are closer to the center move faster. The mass distribution
of the solar system is similar to the rotation curve. Since most of the mass of
the solar system (99.8%) is contained within the central body (the Sun), the
mass distribution shows a rapid decrease with increasing distance from the
Sun.
It should be clear at this point that the mass distribution of the rotating
system gives some hint about the rotation curve of the system. In both cases
examined, the rotation curve and the mass distribution curves match up very
well. So, the rotation curve of a galaxy could be used to determine the mass
and mass distribution of that galaxy.
Galaxies are more like the solar system than like a solid body, since they are
made up of many parts that are all moving around a gravitational potential
well (the galactic center). Astronomers did not know the mass distribution of
a galaxy, but it could be surmised by considering the light curve of the galaxy.
If one makes the assumption that everything that contributes to the mass of
the galaxy emits light, then one can observe the light and map its intensity
Arms in Disk Galaxies • 155
Figure 7.5
The graph shows a light distribution curve for a galaxy. Notice that the light
decreases exponentially as the distance from the center increases. This means that as you
go farther out from the center, the mass decreases exponentially. [Jeff Dixon]
in relation to the distance from the center of the galaxy to determine the
mass distribution of the galaxy. Since all the matter we know to be contained
within galaxies does emit light, this seems like a good assumption to make.
Looking at the light distribution of any spiral galaxy (see example below),
it is easy to see that the implied mass distribution would indicate that the galaxy
rotation curve should look a lot like the rotation curve of the solar system.
What was observed was completely unexpected and did not match either
of the two models considered. When it was observed in a few galaxies,
astronomers thought it might be a strange anomaly, but when it appeared to
be pervasive and most, rather than some, galaxies exhibited this behavior, it
called into question what we thought we knew about the universe—namely
the law of universal gravitation.
So, what did astronomers find? They found that galaxies do not rotate like
solid bodies, nor do they follow a Keplerian rotation curve. Rather, galax-
ies rotate differentially. This means that the stars, gas, and dust in the galaxy
rotate so that the amount of time it takes for material at any given radius to
complete one full rotation is different than at any other radius.
What does it mean that the rotation curve does not match the pre-
dicted mass distribution? Well, it means one of two things: either the mass
distribution is wrong (i.e., light does not trace mass), or the rotation curve
156 • STARS AND GALAXIES
Figure 7.6
The graph shows a rotation curve for a galaxy. Notice that, while there is a steep
rise from the center of the galaxy to a point that marks the edge of the bulge (or central)
region, the material within the disk closer to the center moves at approximately the
same rate as the material in the disk farther from the center. [Jeff Dixon]
indicates that some force, other than gravity, is working to keep the galaxy
together.
The rotation curves of galaxies puts into question the assumption that New-
tonian mechanics can explain the motions of everything in the universe.
Astronomy and the study of the universe starts with the basic assumption that
the forces governing any observed motion follow Newton’s laws. Astronomers
cannot go to different parts of the universe and measure the gravitational
constant to make sure it is the same everywhere in the universe, they have
to work on the assumption that it is the same everywhere in the universe.
If all the matter in the universe is visible, then the distribution of mass of
a galaxy should look like its rotation curve, but they are almost opposite one
another. This cannot be explained by Newtonian mechanics, but no other set
of laws of motion seem to be able to explain what is observed. Either we can
use Newtonian mechanics to calculate a mass distribution based on the rota-
tion curve observed, or we assume that the mass distribution is the same as
the rotation curve speed distribution and throw out Newtonian mechanics
in this case.
Astronomers decided that Newtonian mechanics was too helpful to throw
away, but that the data were too convincing to ignore. The compromise is that
Arms in Disk Galaxies • 157
the mass distribution is not what astronomers expected. In fact, the rotation
curves indicate that there is matter in galaxies that does not emit any light.
This matter is now called “dark matter.”
So, the proposal is that the data are correct and that Newton’s laws still hold.
This implies that there is matter that does not radiate. This matter must be
very strange matter. It cannot simply be normal matter, or can it?
Over the years there have been many proposals for the composition of
dark matter that explain the observations. The two most long-lived candi-
dates are massive compact halo objects (MACHOs) and weakly interacting
massive particles (WIMPs).
MACHOs can be anything from black dwarfs (the cold, dead remnant cores
of intermediate-mass stars like our Sun) to black holes (the remnants of
massive stars). Such massive compact objects would not radiate any electro-
magnetic radiation and, probably, exist in fairly large numbers in the halo
(spheroidal component) of galaxies like the Milky Way. These objects should
exist in the halos of spiral galaxies because the halo is one of the oldest parts
of a spiral galaxy, so the stars that are there are ones that formed near the
beginning of the time that the galaxy itself formed. Stars of a wide range of
masses would have formed at that time, and the remnants of those that have
already ceased to exist would be black holes and black dwarfs.
In the last few decades, astronomers have found evidence to support the
hypothesis that MACHOs exist. A group of astronomers have been looking for
evidence of MACHOs by looking for secondary evidence that they exist. Since
MACHOs, by their definition, do not radiate any electromagnetic radiation,
the only way to detect a MACHO is to observe its interaction with an object
in the halo that does radiate electromagnetic radiation. The clever experiment
dreamed up to test this hypothesis was to look for microlensing events.
Microlensing is the process of a massive, dark object altering the shape of
space-time and causing light from a more distant object to be amplified. This
is a phenomenon of gravitational lensing.
Gravitational Lensing
Gravitational lensing is a phenomenon predicted by Einstein in his general theory of relativity. In
his theory, he depicted the universe as a four-dimensional entity that has three dimensions of space
and one dimension of time. To visualize this, think of a sheet that can stretch. The sheet represents
space-time. A massive object, like the Sun, would be depicted as a bowling ball, for example, on this
158 • STARS AND GALAXIES
stretchable sheet. The mass of the Sun would stretch the sheet. In this way, Einstein talked about
massive objects stretching space-time. Since everything, including light, can only travel along the
sheet, the presence of the massive object changes the way the light will move.
We can observe the change in the shape of space-time by noting that the planets in our solar
system are stuck orbiting our Sun because they cannot get out of the dip in the sheet of space-time
that our Sun has created (sometimes called a gravitational potential well). What our understanding
of space-time did not predict, but Einstein’s did, was the bending of light due to the presence of the
Sun. Einstein’s understanding of the universe predicted that the light from the stars behind the Sun
would be bent as they passed around the Sun to be viewed on Earth. So, during an eclipse of the Sun,
astronomers carefully measured the positions of the stars nearest the Sun and found that Einstein
was correct. The light from these stars did, indeed, appear to have been bent in exactly the way Ein-
stein predicted. (The stars were not in exactly the place they should have been, had their light passed
by the Sun without being bent.) Further, it is possible to see the light of distant galaxies bent around
more nearby galaxies all over the universe.
The effect of a gravitational lens increases the amount of light that reaches the observer. It does
not change the color of the light or the distribution of the light across the spectrum, so one would
still be able to accurately measure a red shift and a blackbody temperature from the light of a gravi-
tationally lensed object. What makes this brightening of light distinguishable from the brighten-
ing observed in a variable star is the smoothness of the light curve and the sharpness of the peak
intensity.
WIMPs is another way to explain the flat rotation curves of spiral galaxies.
Perhaps, particles like neutrinos have a mass that is large enough, that when
added up, accounts for the mass distribution indicated by the light curve. We
know, for example, that neutrinos are created by the billions within the core
of every star in the Galaxy and that they travel for great distances without
interacting with anything. Further, billions of neutrinos are also created dur-
ing supernova events. So, spiral galaxies, like the Milky Way should be full
of neutrinos.
Neutrinos are extremely low-mass particles. The first detection of a neu-
trino was in 1953. This detection was of an electron neutrino. There are three
kinds of neutrinos, each one is named after the lepton particle associated
with it. There are muon neutrinos and tau neutrinos. The tau neutrinos are
the most massive, followed by the muon neutrinos, and then the electron
neutrinos.
Arms in Disk Galaxies • 159
n → p + e- + νe
p → n + e+ + νe
Neutrino Detectors
Neutrino detectors are many and varied in their methods to detect the elusive neutrino. Most are
scintillation detectors, which means they detect something called “Cerenkov radiation,” which is
produced when neutrinos interact with other particles. This radiation is due to the fact that the par-
ticles produced move faster than the speed of light through the scintillating medium.
160 • STARS AND GALAXIES
The Super Kamiokande is one of the most well-known neutrino detectors located 1,000 meters
below ground in the Mozumi Mine in the Kamioka area of Hida, Japan. This detector is simply a
huge tank (50,000 tons) of ultra-pure water Surrounding the water are over 11,000 photomultiplier
tubes that can detect the very faint light of Cerenkov radiation that is produced when neutrinos react
with the water.
A similar detector, known as the Sudbury Neutrino Observatory, located about 2,000 meters
underground in the Creighton Mine in Sudbury, Ontario, Canada, uses what is known as “heavy”
water as its medium. Heavy water is just water where the hydrogen atoms are replaced with deute-
rium atoms. Since deuterium is just hydrogen with an extra neutron, the water molecules are more
massive than water made with the more common isotope of hydrogen; hence, the name heavy water.
Using heavy water in the neutrino detector has the advantage of allowing the detection of muon and
tau neutrinos, instead of just electron neutrinos.
Another type of detector, like the one near Leads, South Dakota, contain a large amount (470
metric tons) of chlorine-containing fluid, like tetrachloroethylene. When a neutrino interacts with
the chlorine it changes it into argon. The argon can be removed and used to measure the number
of neutrino detections. The detector in Leads was the first to measure the solar neutrino deficit. A
similar design uses gallium, which is turned into germanium when it interacts with neutrinos.
So there are billions of neutrinos coming out of stars, like our Sun. Still,
the neutrino mass is so small, even when we add up billions of billions of
them, we still don’t get enough mass to explain the flat rotation curves of
spiral galaxies.
At this point in time, astronomers agree that it is likely that some combi-
nation of MACHOs and WIMPs must account for the flat rotation curves of
spiral galaxies. The idea of searching for other types of matter (like axions)
has been passed over at this point in time. Astronomers seem relatively sat-
isfied with MACHOs and WIMPs as the majority contributors to the dark
matter.
RECOMMENDED READINGS
Mergers and close encounters are the most violent interactions between gal-
axies. A merger is when two galaxies interact gravitationally and only one
galaxy is left at the end. In this scenario, the two galaxies that interacted have
merged to become one, new galaxy. A close encounter, on the other hand, is
when two galaxies interact and get very close to one another, but remain two
separate galaxies. A merger may be preceded by several close encounters.
Galaxy interactions occur on enormously long timescales. No human
has ever witnessed a galaxy interaction in its entirety. We see snapshots of
galaxy interactions in various stages and can surmise the conditions of the
161
162 • STARS AND GALAXIES
Figure 8.1 This diagram shows how the chance alignment of two galaxies can make it appear that they
are interacting, when, in fact, they are hundreds of kpc apart. Some compact groups have chance
alignments, rather than physical interactions. Others may contain both chance alignments and physi-
cal interactions, while still others are entirely physically interacting. [Jeff Dixon]
of galaxies than the number of galaxies. Clusters have more galaxies per unit
volume than groups do. Compact groups have a small number of galaxies
(usually less than 10) that appear to be so close to one another that their
densities (number of galaxies per unit volume) approach those of clusters.
Compact groups are interesting to astronomers because some of the “snap-
shots” astronomers have that show galaxies interacting with one another are
from compact groups of galaxies. While many compact groups of galaxies
are physical groups that are gravitationally interacting with one another,
some are just chance alignments. That is to say that some of the galaxies in
compact groups are not really as close together as they appear to be. Instead
they may be very far apart in radial distance from Earth.
Galaxies are typically about 20–30 kiloparsecs (kpc) in diameter, and are
typically separated by about 1 megaparsec (the equivalent of 1,000 kpc). In
compact groups, galaxies appear to be separated by less than a galaxy diam-
eter; chance alignment, however, can make galaxies appear to be much closer
together than they actually are. Galaxies could be separated by as much as
100 kpc in radial distance, but because their separation is along our line of
sight, they look like they are separated by less than one galaxy radius.
The compact groups that are physically interacting are good laboratories
for understanding galaxy interactions. The galaxies in these compact groups
are very close together and are gravitationally bound to one another. Because
the galaxies are so close together, their interactions are violent and produce
interesting features that allow astronomers to test some of their ideas about
galaxy interactions. It stands to reason that compact groups are the most
likely sites of galaxy interactions.
PAIRS OF GALAXIES
The simplest case of galaxies interacting is with pairs of galaxies. While the
most interesting interacting galaxies are in compact groups, the most preva-
lent modeling has been done for pairs of galaxies. The obvious reason for this
164 • STARS AND GALAXIES
is that the programming of more than two galaxies takes a huge amount of
computing power, which simply does not yet exist.
There are several interesting and well-studied interacting pairs of galaxies.
Models showing how these pairs formed their present features have been
largely successful in that they reproduce the observed structures without
much in the way of assumptions about mass, speed, angle of interaction, or
halo (dark matter) mass. When structure can be reproduced using measur-
able and measured information, the modeling is considered successful. The
most successful models have been used to further predict the unobserved
(or, actually, the as-yet-unobserved).
Unlike stars, galaxies appear not to have formed in clusters, but to have
clustered together over time. This clustering effect will be discussed in greater
detail in chapter 10. What is most interesting about studying interacting
galaxies is that while there are clearly some galaxies that formed from the
beginning as elliptical galaxies, it appears that many formed as the result of a
merge event between two galaxies. In fact, the conditions are most favorable
for producing an elliptical galaxy from a merging pair, no matter what the
morphological type of the galaxy before the merger.
It turns out that it is much less likely that a spiral galaxy will be the result
of a galaxy merger than that an elliptical galaxy will be the result. To get any-
thing other than an elliptical galaxy, you have to start out with two spirals
whose disks are aligned to one another during the merger event. In addition,
the two disks must be rotating so that when they merge, the material in each
disk simply becomes a part of the resultant disk, so they would have to be
rotating with the same speed to not disrupt the disk completely.
Knowing this, the fact that there are so many spirals (about half the galax-
ies in the universe) is a puzzle. Either the probabilities of the outcomes of the
interactions that scientists have come to is incorrect, or most galaxies have
not experienced a merge event. Of course, the latter is the most likely answer.
As the universe expands, the space between galaxies increases, making it less
and less likely that galaxies will interact, much less merge. In the present uni-
verse, we see very few merger events, but many interacting pairs and groups
of galaxies. And, in the early universe we see far more interacting and even
merging pairs of galaxies; however, even in the early universe, not all the gal-
axies are merging, or even interacting. The fraction of galaxies interacting in
the early universe was much greater than in the present universe. Addition-
ally, the fraction of interacting galaxies that are merging was also greater.
This is largely because galaxies were closer together in the early universe, so
they interacted more. In addition to this, the objects that merged to become
galaxies may have formed in regions of over-density. This means that galaxy
seeds were already in clusters so that they would interact and merge and
form present-day galaxies. This also explains why so many galaxies in the
present universe are members of clusters or groups.
When galaxies interact or merge, the gas within them is used to form stars.
Such bursts of star formation occur in the stellar history of every elliptical
Galaxy Interactions • 165
galaxy. One can use models to predict the color, spectral energy distribution,
and gas content of galaxies, given a particular star formation history. Doing
this, astronomers have concluded that every elliptical galaxy is an elliptical
galaxy because near the beginning of its existence, it experienced a large burst
of star formation. As such, a majority of stars in the galaxy, now, are near the
ends of their evolutionary sequences. The galaxy appears red because most
stars are red giants or red dwarfs.
The spectral energy distribution indicates that most of the energy com-
ing from elliptical galaxies is at longer wavelengths, indicating that the stars
in these galaxies are emitting low-energy photons, so they have low surface
temperatures and are mostly on the right-hand side of the H-R Diagram,
indicating that they are evolved (red giants) or very low mass (red dwarfs).
Also, elliptical galaxies have little or no gas and dust in them. This is con-
sistent with the proposed scenario of a burst of star formation. Such a burst
would use up all available gas and dust in the formation of stars.
Since all elliptical galaxies fit this model, it is not inconsistent with the pro-
posed hypothesis that all elliptical galaxies are the results of merger events
that occurred early in the history of the universe. This is not proof that all
elliptical galaxies formed this way, but recent observations of early universe
are beginning to show evidence that supports this hypothesis.
found on the Web. The prediction is made considering the initial speeds of
the two galaxies as well as the mass of the two galaxies and the inferred dark
matter halo mass. Recent observations put the dark matter halo masses of
Andromeda and the Milky Way at roughly 1012 M~ (solar masses) each. This
is approximately equal to the mass of each galaxy, so that the dark matter
halo accounts for about half the mass of each galaxy.
Astronomers predict that the Milky Way and Andromeda will merge dur-
ing this collision. Most of the gas and dust within both galaxies, if not all of
it, will be used to form a new generation of stars during the merging process.
What will happen to Earth and our solar system? Probably, not much will
change. It does depend on where the Sun is, in relation to gas clouds, during
the collision, but most likely the Sun will be moved around, relative to the
center of the Milky Way, during the merging process. The Sun could end up
closer or farther from the galactic center.
Barred spiral galaxies are spiral galaxies whose bulge is extended along one
axis, making it look more like a bar than a sphere. The formation of such a
feature both causes and is caused by a gravitational disturbance. To elongate
the spheroidal bulge requires some kind of gravitational perturbation. While
astronomers have not been able to single out a particular mechanism that
accounts for all bar formation, it is clear that at least some bars are due to
galaxy interactions. It is also clear that the formation of a bar at the center of a
galaxy causes perturbations in the disk of that galaxy, which may cause inner
or outer rings to form, as well as forcing a two-arm spiral pattern.
A bar at the center of a disk is very different from a spheroidal bulge at the
center of a disk. First, it changes the look of the gravitational potential well. A
spheroidal bulge has a gravitational potential well that look like a cone. How-
ever, a bar makes the point of the cone wider. To complicate things more,
the bar rotates, and this rotation twists the wide point of the cone. So, large
bars can greatly affect the shape of the gravitational potential well of a disk
galaxy.
The gravitational potential well plays a significant role in the development
of spiral density waves. It is the framework on which the spiral density waves
are laid. Depending on how strongly the bar affects the underlying gravita-
tional potential, a bar can drive spiral density waves. Conversely, a small or
weak bar can be present in a disk that has no spiral density waves.
The stronger and larger the bar, the more grand design the spiral arm
pattern becomes. Large, strong bars can also extend beyond the radius of
Galaxy Interactions • 167
corotation. In disk galaxies with bars, there are two different motions: the
motion of the bar feature and the motion of the disk. Just as the spiral den-
sity wave pattern motion is different from disk rotation, the bar feature and
the disk motion are separate. The bar feature rotates as a solid body in this
case, so the stars appear to move faster with increasing distance from the
center point; the disk, however, rotates differentially, so the gas and dust
and stars move at the same rate, no matter how far they are from the center.
The radius of corotation is where the disk and bar feature are rotating at the
same rate.
While the stars are orbiting the center of the galaxy, they are not orbiting
in circular, or even elliptical, orbits. The orbits of the stars within a galaxy are
more complex. They involve a lot of perturbations (due to the gravitational
potential well as well as gravitational interactions with other stars orbiting
the galaxy). Most stars appear to orbit the galaxy in a “wobbly” pattern due
to their orbits having epicycles.
Generally, if the radius of corotation is close to the location of a resonance
(when the difference between the speed of the stars and the pattern speed of
the bar or the disk is equal to a whole number of epicycles), it marks the loca-
tion for an inner ring feature. Galaxies can have inner rings, nuclear rings,
and outer rings. This ring structure is usually an indication of resonances
that occur within a galaxy. Galaxies with strong resonances often are home
to spiral density waves.
Resonances can also occur within a bar. Stellar orbits within a bar can vary
widely, causing different kinds of bars. These shapes may or may not indicate
the presence of density waves or the strength or size of the bar. As a result of
the differing possible stellar orbits, bars can be long and narrow or wide and
boxy or even, peanut shaped.
The presence of a bar in a disk galaxy with an existing spiral density
wave system can cause multiple symmetries. That is, there may be a two-
arm pattern near the center that changes to a three-arm pattern after
corotation. Many unexpected symmetries have been discovered in spiral
galaxies with bars and without bars. Multiple symmetries can occur when
two competing spiral density waves are present, which does not require a
bar, per se.
While there are galaxies that have bars and are flocculent (without spi-
ral density waves), the bars in these galaxies do not extend to corotation.
Most galaxies with bars that extend to corotation have a grand design arm
pattern. Dynamically, this can be explained by the gravitational potential of
the galaxy being dominated by the bar causing a perturbation within the
disk, which results in a spiral density wave. Observations so far support this
hypothetical scenario. However, not every galaxy in the universe has been
assigned a spiral arm class, or classified by its bulge shape. So, while astrono-
mers seem to be in agreement that this explains what is currently observed,
the explanation is contingent upon what is observed. If new observations are
counter-indicative of this hypothesis, then it will be altered.
168 • STARS AND GALAXIES
Bars are not observed in every spiral galaxy in the universe; however, most
spiral galaxies contain bars. This fact alone is indicative that bars must be
transient, but long-lived features. That is, bars must be able to exist and then
disappear. Also, bars must be able to appear where they were not before.
One mechanism for the creation of bars is galaxy interactions. When galaxies
interact, their gravitational potential wells are disturbed. A bar is a perturba-
tion, or disturbance, within a gravitational potential well. Therefore, astrono-
mers think it is possible that bars can be caused by galaxy interactions. This
would explain the prevalence of bars in the universe, while allowing for the
existence of a significant population of galaxies without bars.
There is some observational evidence to support this. Many barred galax-
ies show evidence of galaxy interactions. Some show evidence because they
have nearby neighboring galaxies that may be currently interacting gravita-
tionally. Some show evidence because they have had past epochs of rapid star
formation. However, not every barred spiral galaxy has obvious evidence of
galaxy interactions.
A possible mechanism for destroying and re-creating bars is active galac-
tic nuclei (AGN), which are thought to be supermassive black holes found
in the centers of some galaxies (both spiral and elliptical galaxies can have
AGN). These supermassive black holes have accretion disks around them,
which can extend to as far as several parsecs from the black hole. The accre-
tion disk is just a disk of material that is being accreted by the black hole.
Because of its mass (usually equal to or greater than 106 M~), the black hole
can attract matter to its accretion disk from great distances.
A bar feature works to fuel an AGN because it drives material towards the
AGN. As it does so, it removes the bar feature by removing the gas clouds
in the orbits that create the bar feature. It is mainly the inflow of gas, not
stars that destroys the bar feature. Slowly, the inflow of gas causes the bar
feature to bulge out, becoming more lens-like, and, eventually, creating a
bulge. Since the AGN mainly consumes gas, it is this inflow that destroys the
bar. However, once the bar is destroyed the inflow of gas stops, which shuts
down the AGN, allowing the bar to re-form.
More detail about how AGN function will come in chapter 9; however, in
a nutshell, the supermassive black holes that fuel AGN are thought to be the
results of galaxy mergers. It is not clear whether supermassive black holes
were formed before their host galaxies or vice versa. There are many objects
found in the early universe that contain supermassive black holes and do not
appear to have undergone any interactions or mergers, and yet almost every
AGN in the present universe shows some evidence of a past interaction or
merger event. It is not clear what role the supermassive black holes found
at the cores of some galaxies play in galaxy evolution. Much of the answer
to this puzzle comes from understanding how galaxies formed and evolved.
This is the topic of chapter 10.
Galaxy Interactions • 169
RECOMMENDED READINGS
SEYFERT GALAXIES
Seyfert galaxies are a class of galaxies named for their discoverer, Carl Seyfert.
Carl Seyfert discovered this class of galaxies in 1943. They are described as
a class of spiral galaxies whose nuclear spectra have unusual emission lines.
Since the centers of spiral galaxies contain mostly stars and very little gas,
the spectrum of a spiral center should be an absorption spectrum matching
the spectral type of mostly evolved stars. However, Seyfert found that there
were many spiral galaxies that had emission spectra, not absorption spectra.
This type of spectrum indicates the presence of hot, ionized, gas. Even more
interesting, this hot, ionized gas found in the cores of Seyfert galaxies is not
associated with the kinds of stars that are usually associated with ionized
gas. There were actually two different types of emission spectra that Seyfert
observed, which is why there are two kinds of Seyfert galaxies. These sub-
classes were not identified until 1974, when higher resolution spectra were
able to pick up these subtle features.
171
172 • STARS AND GALAXIES
Seyfert Type 1
Seyfert 1 galaxies are galaxies whose nuclei have spectra with both broad
and narrow emission lines. The broad lines are a result of the Doppler effect.
Because the gas is moving very fast (usually thousands of kilometers per
second), the wavelength of the light emitted is shifted. Since the gas in the
nucleus of a Seyfert galaxy is moving in an accretion disk, sometimes it is
moving towards Earth and sometimes it is moving away from Earth. When
the gas is moving towards Earth, the light it emits is shifted to the blue end of
the spectrum. When the gas is moving away from Earth, the light it emits is
shifted to the red end of the spectrum. So, since we are observing the entire
disk, some of the light is shifted to the blue end of the spectrum and some of
the light is shifted to the red end of the spectrum. As a result, we see a line
that is very wide extending from a much shorter wavelength than the emitted
light to a much longer wavelength than the emitted light.
Also, the width of the emission lines indicates the presence of different
densities of gas. The broader the emission lines, the denser the gas. This is
because the broader lines are indicative of higher velocities for the particles
within the gas that is emitting light. The faster moving gas, must be more
dense in order to stick together at such high speeds. As a result of these
observations of broad line emission spectra, astronomers have concluded
that Seyfert 1 galaxies are spiral galaxies that have supermassive black holes
at their cores that are accreting material from an accretion disk, which is
emitting light as it rotates and drops material on the central supermassive
black hole.
The spectra of Seyfert 1 galaxies indicate that in the nuclei of these galax-
ies there is both high-density and low-density gas emitting light. The high-
density gas is in the accretion disk, because it is rotating. The low-density
gas is not in the accretion disk, rather, it is in a torus (donut-shaped ring) of
gas that surrounds the accretion disk. This is the gas that produces the nar-
row lines. This gas is not moving as fast as the denser gas, so the emission
lines from this region are not as broad. The narrow-line producing gas is also
thought to be associated with a ring of star formation that may be the result of
the inflow of gas to the accretion disk to fuel the supermassive black hole. The
inflow of gas to the accretion disk causes gas clouds to collide, which causes
star formation, so surrounding the accretion disk is a ring of star formation.
This ring of star formation is the source of the narrow emission lines.
Seyfert Type 2
Seyfert 2 galaxies do not exhibit the broad lines in their spectra. Rather,
they appear to only have narrow emission lines at their cores. Since the cen-
tral engine for both Seyfert 1 and Seyfert 2 galaxies is the same, astrono-
mers think the reason that the broad emission lines are not seen in Seyfert
Active Galaxies • 173
RADIO GALAXIES
Radio galaxies are galaxies that emit most of their electromagnetic radiation
in the radio regime of the electromagnetic spectrum. Often, they have a vis-
ible jet, or two jets that are symmetrical. These jets are associated with large
radio lobes that extend much farther than the galaxy itself. The majority of
the radio radiation from these radio lobes is synchrotron radiation, indicat-
ing that the material in the lobes contains charged particles and a strong
magnetic field.
In addition to the radio radiation, the lobes, jets, and cores of these galax-
ies emit a lot of x-ray radiation. The recent x-ray satellites have added much
to our understanding of radio galaxies because of their ability to detect and
map the locations of the x-ray radiation associated with these objects. X-ray
radiation is indicative of a process that is capable of producing such high-
energy radiation. The lobe and jet features are also indicative of the nature
of the compact energy source at the cores of these galaxies. A supermas-
sive black hole with an accretion disk that is actively accreting material can
explain the existence of the jets, the radio lobes, and the x-ray radiation.
The mechanisms for creating these features are fairly straightforward,
in the scheme of things. Material accreting onto the black hole by way of
the accretion disk gets heated up to extremely high temperatures. This
extremely hot material radiates x-ray radiation. Within the accretion disk,
material is experiencing extremely strong forces. The gravitational pull of the
supermassive black hole is tearing material apart because the strength of the
174 • STARS AND GALAXIES
force is so great that tidal forces are strong enough to tear apart atoms. So,
as the material is accreted, it is broken down into its smallest pieces freeing
electrons within the disk, which will move along established magnetic field
lines creating synchrotron radio radiation.
Tidal Forces
Tidal forces are forces that are caused by a large difference between the strength of a force on the near
and far sides of a solid object. For example, the gravitational force of Jupiter, at very close distances
from Jupiter, decreases significantly with distance. That is, at the distance of the average orbit of Io,
its nearest satellite, Jupiter pulls much more on the near side of Io than on the far side of Io. That
difference is large enough to stretch Io and cause it to be extended along the line of the gravitational
pull of Jupiter. As a result, Io is alternatively stretched and squished throughout its orbit of Jupiter.
This keeps the core of Io molten and makes the surface of Io active and young in appearance. This is
how tidal forces can work to tear apart solid objects. The more massive the gravitational source, the
greater the difference in force strength at distances close to the object.
So, the tidal forces caused by being very near a supermassive black hole are strong enough to tear
apart atoms. This is significant because, comparatively, the gravitational force is 38 orders of magni-
tude weaker than the nuclear force, which holds the nuclei of atoms together; however, supermassive
black holes are unusual objects around which rare things do occur.
Event Horizon
A black hole is point in space known as a singularity. The object itself is dimensionless. A black hole
is a point of mass and energy. It is said that even light cannot escape a black hole. In itself, this last
statement is not entirely an accurate description of a black hole. The mass of a black hole is so com-
pact that there is a relatively large area around a black hole where the escape velocity is greater than
the speed of light. This area is called the event horizon.
The size of the event horizon of a black hole depends on its mass. A small black hole will have
a small event horizon, but a massive black hole will have a larger event horizon. The event horizon
describes the area around a black hole where the escape velocity exceeds the speed of light. Once an
object crosses the event horizon, it cannot escape the gravity of the black hole and it is doomed to
become part of the black hole.
Active Galaxies • 175
Radio galaxies are mostly elliptical galaxies. Many of them are giant ellip-
tical galaxies. Giant elliptical galaxies are located at the centers of large clus-
ters of galaxies. Presumably, these giant elliptical galaxies are the results of
multiple merger events. Sometimes giant elliptical galaxies are described as
cannibal galaxies because they lie at the bottom of the deep gravitational
potential well of the dense galaxy cluster in which it sits and all the other
galaxies in the cluster are moving towards or interacting gravitationally with
the giant elliptical galaxy.
QUASARS
The word “quasar” is short word for the actual name of this object: quasi-
stellar radio source. When first discovered, it was not clear what the nature
of these objects was. They appeared to be point-like sources of light, so they
were called quasi-stellar objects, or QSOs. The reason they were not consid-
ered stars is because their spectra were redshifted as much as (or more than)
the most distant galaxies, and yet they were much brighter than these distant
galaxies. Since there were not very many of these objects known at first, and
there were no models or theories that could explain how they could produce
so much energy and still appear to be point sources.
As observations increased the number of quasars known, astronomers
began to develop models for how so much energy could be produced in such
a small object. Debate grew about whether the redshifts of quasars meant
the same thing as the redshifts of galaxies. Perhaps there was another way to
cause a redshift that wouldn’t imply that these objects were so distant. The
great distance of quasars presented a problem only in that if quasars were as
far away as their redshifts indicated, they were the brightest objects in the
universe—brighter than any known objects.
Eventually, a handful of quasars were found to be hosted by spiral galaxies.
These observations were made with the newly repaired Hubble Space Tele-
scope and with the extremely large ground-based telescopes. Now, astrono-
mers can show by spectroscopic analysis and high resolution imagery that
quasars are the bright nuclei of very distant galaxies.
At this point in time, it is commonly accepted that quasars are an early
universe type of AGN. Conditions in the cores of these galaxies are differ-
ent than those in the galaxies we see that are nearer to the Milky Way. In
the cores of galaxies that host quasars, there is more gas than in the cores
of nearby galaxies. If the mechanism for producing energy in quasars is the
same as in nearby AGN, there is much more fuel for the fire, so to speak, in
the quasar phenomenon than in either the radio galaxy or the Seyfert phe-
nomenon. This is why quasars are so much brighter than their nearby AGN
counterparts.
The other observable trait of quasars is the presence of superluminal ejecta.
Several quasars have been observed to eject material (observed in the radio).
176 • STARS AND GALAXIES
Over the years, astronomers have been able to observe the ejected material
moving away from the quasar source. The rate at which these ejecta move
away from the quasars is faster than the speed of light. Obviously, the mate-
rial cannot really be moving faster than the speed of light, but simple calcula-
tions of the rate of speed of these objects does result in speeds greater than c.
The effect that allows these ejecta to appear to move at superluminal speeds
is a relativistic effect called Doppler boosting.
Doppler Boosting
The universe is expanding at a rate of about 72 km/s/Mpc. That means that an object that is 1 mega-
parsec distant is increasing in distance by 72 km every second. That is not very much, since a mega-
parsec is 3.086 x 1019 km. So in one second an object that is 1 megaparsec distant goes from being
30,860,000,000,000,000,000 km away to being 30,860,000,000,000,000,072 km away. At large dis-
tances, this can become significant. The distance at which most quasars are found is greater than
1,000 Mpc, so the change in length of the linear distance is 72,000 km every second. An object at this
distance that is moving close to the speed of light (299,792 km/s) will appear to be moving faster than
the speed of light because of the rate of expansion of the space between the observer and the object.
This object will appear to have superluminal speed.
Doppler boosting is an artifact of the expansion of the universe. When viewing objects that are
at great distances from Earth, the motion of the expansion of the universe is the bulk of the motion
observed. For objects that are moving very fast (near the speed of light) the expansion of the universe
adds motion that can make the object appear to have superluminal motion. However, no object can
move at a speed that exceeds the speed of light, so this is just an apparent motion.
Radio and x-ray observations of these objects show that they are sources
of high-energy photons and synchrotron radiation. The observed super-
luminal motions are indicative of jets that appear to exhibit superluminal
motion because of Doppler boosting, as in quasars. Yet, these objects are
different from quasars in that their spectra do not indicate the presence of
gas or dust.
In the core of each of these types of AGN lies a supermassive black hole
with an accretion disk and a torus of dust surrounding it. This is known
as the central engine. The characteristics of this central engine are empiri-
cally derived. That is, observations have led to these conclusions. Astrono-
mers have observed similar energy amounts and energy distributions from
these AGN.
Each source is contained in a very small space or exhibits variations over
very short time scales (minutes or seconds). The small size of the central
engine is derived from this fact. Since nothing can change faster than the
speed of light, the variations in brightness observed in these different types
of AGN limit the possible size of the central energy source. Variations in
intensity that occur in minutes imply sizes not much bigger than our solar
system.
The enormous amounts of energy observed coming from these extremely
small sources makes these objects extremely interesting. Astronomers have
developed a model of a system that can explain all the observations of AGN
so far. Different AGN appear different due to their environments and orien-
tations. The current Unified Model explains every observation to date. New
observations are used to test the model. If something is observed that cannot
be explained by the model, it will be altered until it can explain the observed
phenomenon. It is approaching the theory state, but there is still much more
to learn about supermassive black holes and their behavior before the Uni-
fied Model can graduate to theory.
The Unified Model was first proposed in the 1980s. Astronomers began to
notice that these four classes of galaxies had much in common. They all
exhibited high-energy x-ray radiation, variability over short timescales, syn-
chrotron radio radiation and radio jets, lobes or ejecta. So the energy source
of these objects must be massive and small.
Given the amounts of energy released and small timescales over which
variations are observed, the size of the energy source must be similar to the
size of the solar system. The only type of object that could be so small and so
178 • STARS AND GALAXIES
Figure 9.1The schematic above depicts the current model of the central engine that pow-
ers all active galactic nuclei (AGN). At the center is the supermassive black hole. Sur-
rounding that is an accretion disk of material. Around that is a torus of dust and gas.
Perpendicular to the accretion disk, jets and radio lobes can be observed. Depending
on the angle at which the lobes are viewed, they may both be seen, only one, or none.
[Jeff Dixon]
Active Galaxies • 179
star formation and serving as a mechanism for re-starting the central engine
when the accretion disk runs out of material.
Astronomers have created this model based on all the observations for
all the different types of active galaxies observed in the universe. Th e model
was first conceived to explain the difference between Seyfert 1 and Seyfert
2 galaxies. Astronomers who observed radio galaxies, quasars, and BL Lac
objects began to see that this proposed energy source could explain their
observations as well. Interestingly, the connection of this energy source for
all active galaxies further enhanced the attempt to put together galaxy evo-
lution. Since quasars and BL Lac objects are high-redshift, early universe
objects, and since Seyferts and radio galaxies are low-redshift, recent-universe
objects, the fact that one energy source model can explain all active galax-
ies means that such an energy source is an important connection between
early-universe objects and objects in the current epoch. Galaxy evolution
and how active galaxies have helped piece that together is the topic the next
and final chapter.
RECOMMENDED READINGS
Like stellar evolution, the evolution of galaxies is not something that humans
can observe since it takes much longer than a human lifetime for a galaxy
to change. The fact that galaxies change is evident from observing galaxies
at different epochs in the universe. Looking at all the snapshots of galax-
ies over time, astronomers have pieced together a galaxy evolution scenario.
Additionally, they have considered what is known about the composition of
the universe (the amount of cold dark matter and hot dark matter and dark
energy) to further test and constrain this model.
In this chapter, we will explore the various environments of galaxies, the
idea of look-back time, and what is known about galaxy evolution. Finally,
we will examine the most recent galaxy evolution models and describe what
data they are based on and what future observations will help to constrain
this model and how they will constrain it.
GROUPS
181
182 • STARS AND GALAXIES
Because we live in the Milky Way, there is a large portion of the sky that
is not visible. This is because the Milky Way is blocking our view. For that
reason, we cannot say that astronomers have identified every galaxy in the
Local Group. Recently two large galaxies were discovered that are mem-
bers of the Local Group. Both galaxies are located exactly in the region of
the sky that is obscured by the dusty disk of the Milky Way. Both are fairly
large galaxies compared to the others in the Local Group. So, it would not
be unexpected if smaller galaxies were found that are, by chance, located in
the region of the sky that is blocked from our view by the dusty disk of the
Milky Way.
Besides the Local Group, there are other groups of galaxies, but most of
them are much smaller than the Local Group. Many are also compact groups
of galaxies (discussed in more detail in chapter 8). Most galaxy groups are
named for their central member, such as the M81 group.
CLUSTERS
than the third brightest galaxy within a radius of 1.5 Mpc from the center.
If the count of galaxies of that brightness within that radius is less than 30,
the cluster is considered poor. Groups are also considered a type of poor
cluster.
SUPERCLUSTERS
REDSHIFT
To understand how all these pieces fit into the current understanding of the
universe and galaxy evolution, the concept of redshift is a necessary piece of
information. Astronomers can measure the motion of things in the universe
using redshift. This phenomenon is an aspect of a well-known phenomenon
observed on Earth known as Doppler shift.
Doppler shift is the phenomenon of waves changing wavelength due to the
motion of a source or observer. Doppler shift occurs when a sound source,
like a train, is moving towards an observer. In this case, the sound the train is
making is shifted to a higher pitch as the train approaches. When the source
is moving away from the observer, the sound is shifted to a longer wave-
length, resulting in a deeper pitch as the train recedes. The same phenom-
enon can occur if the observer is moving and not the source. For example,
as an observer approaches a loud carnival, the sounds get higher and higher
in pitch until the observer passes the carnival and the sounds shift to lower
and lower pitches.
In astronomy, we consider that the observer is stationary and the source is
moving, although it really does not matter which one is moving—the source
184 • STARS AND GALAXIES
or the observer. The waves in astronomy are light waves, not sound waves, so
the shift is in color, not pitch. When an object is moving toward the observer,
the light emitted is shifted towards shorter and shorter wavelengths, so the
light appears bluer than it should. This is a blueshift. Conversely, when an
object is moving away from an observer, the light emitted is shifted to longer
and longer wavelengths, so the light appears redder than it should. This is a
redshift.
Because the universe is expanding, the space between Earth and every
galaxy in the universe is being stretched. So most galaxies are redshifted, due
to the Doppler Effect. Some galaxies are not redshifted because they are mov-
ing towards Earth faster than the universe is expanding the distance between
Earth and the galaxy, so their net motion is still towards Earth. This can really
only occur for galaxies that are relatively close to Earth. This is because it
does not take a very great distance for the rate of expansion of the universe to
exceed normal galaxy speeds.
The more the light is redshifted, the faster the object is moving away.
Because of the expansion of the universe, the faster a galaxy is moving away
from Earth, the farther away it must be. Therefore high redshift is the same
as great distance. Redshift is measured in kilometers per second or frac-
tion of the speed of light. That is, an object that is moving away from Earth
at a speed equal to half the speed of light has a redshift of 1.5 x 105 km/s
or 0.5.
Even though nothing in the universe can actually go faster than the speed
of light, some galaxies have redshifts as high as eight, meaning their light is
shifted so much that their speed away is eight times the speed of light. Since
the motion of a galaxy due to the expansion of the universe is only appar-
ent (in that the galaxy is not actually moving, rather the space between the
observer and the galaxy is increasing due to the universe expanding), galax-
ies can have redshifts that are greater than the speed of light.
LOOK-BACK TIME
themselves to you on their 21st birthday. You would receive Abe’s picture
first, but by the time you got it, Abe would be 26 years old. Beth’s picture
would arrive when she was 31, and Carl’s picture would arrive when he was
41. So, who looks youngest for their age? The friend who lives farthest away
is the one who appears much younger than he is.
Galaxies are millions of light years from Earth. This means that the light
we see when we look at other galaxies is light that left that galaxy millions
(or more) of years ago. To see what a distant galaxy looks like at this instant,
we have to wait millions of years. This may seem like a big problem—that
we never see anything as it is now. But, astronomers have thought of this
as a way to see the past. That is, when we observe distant galaxies, we can
see how galaxies looked millions (or billions) of years ago. If we put them
on a distance-time scale, we can try to piece together how galaxies evolve.
So, observing distant galaxies can be like watching a movie of how galaxies
evolve, if you can piece the parts together correctly.
This idea of observing distant objects to look at how things were in
the past is called look-back time. Since more distant objects are younger
(seen as they were longer ago), astronomers can observe what galaxies
looked like when they first formed, if they can see objects that are distant
enough.
revolutionary images, the space telescope pointed to the same part of the sky
and took images at least 342 separate times. Because the images are digital,
astronomers can add them together, so that faint objects appear brighter. The
total exposure time of all the images when added together is a little longer
than 11 days!
In December 1995, the newly launched and recently repaired Hubble Space
Telescope pointed to a part of the sky that contained no bright stars and no
bright galaxies to speak of. The telescope imaged this part of the sky for such
a long time that the faintest objects in the image have an apparent magnitude
of almost 32. This is 16 magnitudes fainter than the human eye can see (that
is about 10 trillion times fainter).
The Hubble Deep Field was the first deep image that showed astronomers
what lurks on the edge of the visible universe. Astronomers saw, for the first
time, the building blocks of galaxies, and were able to confirm that galaxies
most likely formed from the joining together of smaller pieces, rather than
the breaking apart of larger pieces. The smaller pieces appear to be unstruc-
tured and blue in color, relative to their more modern cousins. We know the
small, blue, fuzzy objects found in the Hubble Deep Field are proto-galaxies
because they are far away, which means we are seeing them as they were a
long time ago.
The Hubble Deep Field was such an important datum for astronomy that
it was decided to do another Hubble Deep Field in the southern part of the
sky. This was an important experiment to make sure that what was observed
in the original Hubble Deep Field was not unique, but represented what is
seen in every direction. The southern Hubble Deep Field demonstrated that
there was nothing special about the Hubble Deep Field. The southern Hubble
Deep Field was so important that astronomers thought it would be a great
idea to extend the depth of the Hubble Deep Field. This is how the Hubble
Ultra-Deep Field was conceived.
The Hubble Ultra-Deep Field is a new image taken with two different
cameras on the Hubble Space Telescope. The image took 1 million seconds
of exposure time and is of another relatively empty part of the night sky. This
image was taken simultaneously with two different instruments—one an
imager and one a spectrograph. This way, each object’s spectrum is also avail-
able. For the thousands of objects observed in the Hubble Ultra-Deep Field,
most have been identified. Some of the objects in this image are the youngest
objects ever seen. These objects are so young that the objects identified as
188 • STARS AND GALAXIES
building blocks of galaxies in the Hubble Deep Field are now understood to
be an intermediate stage between the actual building blocks and the galaxies
of the present epoch.
So, what do astronomers now know about the evolution of galaxies? First, it
is clear that galaxies form from smaller pieces. This is what observations of
the most distant objects reveal. This conclusion has implications for under-
standing the universe since it implies that most of the dark matter in the uni-
verse is cold, rather than hot. This means that there should be more objects
like MACHOs (massive compact halo objects) than like massive neutrinos
(weakly interacting massive particles)—WIMPs. Cold dark matter is simply
dark matter that is slow-moving. (Dark matter is discussed in much greater
detail in chapter 7.)
The space-based telescope deep-sky images have also revealed that the for-
mation of galaxies early on was by merging the proto-galaxies observed in the
early universe. This means that galaxy formation is bottom-up process. This
further implies that the formation of clusters and groups of galaxies is a product
of this same process. The current model for cluster formation begins with the
small over-densities observed in the cosmic microwave background radiation.
These seeds are where there may have been slightly more proto-galaxies
than in other parts of the universe. Those proto-galaxies attracted other
proto-galaxies. Eventually the proto-galaxies merged to form galaxies. In the
areas where the seeds were, there were more galaxies than in other areas.
These galaxies attracted other galaxies. Over time, the gravitational attrac-
tion of the seeds to other galaxies created groups and clusters of galaxies. The
galaxies and proto-galaxies that formed these groups and clusters came from
nearby and far away, leaving empty space behind them.
Some astronomers think it is possible that within the great voids in the
universe (where there are few galaxies) there may be proto-galaxies that
never merged (for lack of neighbors) and have evolved in isolation. Studies
of so-called isolated galaxies attempt to explore galaxy evolution in undis-
turbed environments. Since almost everything we call a galaxy must have
been through at least one merger in its existence, isolated galaxies offer a
unique laboratory for studying galaxy evolution.
Another topic in the realm of galaxy evolution is the presence of bars. Bars
appear to be transient structures that are fairly long-lived. Some dynamic
models suggest that bars may vary in strength over time. If bars are a mecha-
nism to fuel a central supermassive black hole, the idea of varying strength
could be consistent with the idea of turning on and turning off a central
engine. But, not all galaxies with bars appear to house an active nucleus.
The difference between elliptical and spiral or disk galaxies appears to
be star formation history. Star formation history may be a way to read the
Galaxy Evolution • 189
galactic evolution story for an individual galaxy. That is, each galaxy may
have its own history that can be told by looking at star formation history
and determining how stars have formed over time (whether looking at an
early burst of star formation with no recent activity or at nearly constant star
formation rates that are fairly low with perhaps some peaks of star formation
that indicate strong interactions or merger events).
Within disk galaxies there are some with prominent spiral density waves
and others that have no evidence of spiral density waves at all. Disk galaxies
appear to vary widely in arm class. Very little is understood about why spiral
density waves are present in some galaxies while not in others. The presence
of spiral density waves, too, may be a transient feature that is related to merg-
ers or galaxy-galaxy interactions.
The theme that should be evident by now is that galaxy mergers are a very
important part of galaxy evolution. Galaxy mergers may drive galaxy evolu-
tion. Studies of galaxies in the process of merging or interacting violently will
lead to a better understanding of how this process changes galaxies. Studies
of galaxies that only rarely interact with other galaxies may lead to a better
understanding of how galaxies can evolve without the aid of merger events.
The evolution that occurs without galaxy-galaxy interactions is probably
driven by stellar evolution and may be a slower process than the changes that
can occur during merger events and other types of galaxy interactions.
Stellar evolution takes millions to billions of years to occur. It took only
about a century to deduce the processes of stellar evolution through obser-
vations of clusters of stars (see chapter 3 for more on star clusters). Gal-
axy evolution takes at least tens of billions of years. It has only be a century
since astronomers have known that galaxies other than the Milky Way exist,
so astronomers may be close to understanding galaxy evolution, or it may be
another century from now before this tale is unraveled. Certainly, our under-
standing of this process has made great gains thanks to the deep images of
the space-based telescopes. With new technology comes greater opportunity
for deeper understanding.
RECOMMENDED READINGS
absolute magnitude (M). A measure of the energy emitted by a star that uses
a logarithmic scale.
absorption spectrum. A type of spectrum that includes almost all wavelengths
of visible light (those excluded are determined by the chemical composition
of the absorbing medium). It looks like a rainbow with dark lines in it.
accretion disk. A disk of material that is orbiting a black hole or other mass
source (like a star); the material in the disk will become a part of the black
hole by spiraling toward it.
active galactic nucleus (AGN). A type of galaxy core that emits a huge amount
of high-energy electromagnetic radiation as well as synchrotron radiation.
apparent magnitude (m). A measure of the energy received from a star that
uses a logarithmic scale.
arcminute. An angular measurement equal to 1/60th of a degree of arc.
arcsecond. An angular measurement equal to 1/60th of an arcminute or
1/3600th of a degree of arc.
asterism. A group of stars that are part of one or more constellations; such
groupings are not recognized as constellations, but are often used to locate
objects in the sky.
astrometry. The use of accurate measures of positions of stars to determine
the rates of motion or positions of stars or other celestial objects.
astronomical unit (AU). A linear distance equal to 1.5 x 1011 m. The average
distance between Earth and the Sun.
asymptotic branch. The stage of red giant phase achieved only by stars mas-
sive enough to fuse carbon in their cores.
band. A part of the electromagnetic spectrum sampled using a filter that
allows only a small range of electromagnetic radiation to pass through it and
blocks all the rest.
barred spiral galaxies. A disk-dominated galaxy with a central bulge that is
elongated or peanut-shaped.
barycenter. The center of mass of a system of multiple objects.
beta decay or proton decay. The process whereby a neutron decays into a proton,
during this process an electron and an electronic anti-neutrino are emitted.
191
192 • GLOSSARY
BL Lac Objects. Objects like the galaxy known as BL Lac. High-redshift AGN
housed primarily in elliptical galaxies.
black dwarf. A theoretical star that would be the final stage of a white
dwarf when the star cools to a temperature of absolute zero and emits no
light.
blackbody. A theoretical object that absorbs all radiation incident upon it
and radiates all wavelengths of light. A hot solid or hot dense gas.
blackbody spectrum. A plot of the intensity vs. wavelength (or frequency)
of the radiation from a blackbody. The range of intensities per wavelength
or frequency of electromagnetic radiation emanating from a hot solid or hot
dense gas.
blazars. High-redshift astronomical objects that house active galactic nuclei
(AGNs) and are associated with elliptical galaxies. (Also known as BL Lac
objects.)
blueshift. The shift of light toward the blue (shorter wavelength) end of the
spectrum due to motion towards a source.
Bohr model. A model of what an atom looks like that is useful for explaining
the phenomena of emission and absorption spectra; the model describes an
atom as a nucleus with electrons that exist only on specific energy states or
orbits.
brightness. The amount of energy received from a light-emitting source.
bulge. The spheroidal component of a disk-dominated galaxy located at the
center; a bulge that is elongated or peanut-shaped is called a bar.
central engine. The power source of an active galactic nucleus. A supermas-
sive black hole with an accretion disk, torus of dust, radio lobes, jets and star
forming regions at the core of an active galaxy.
Cepheid variable. A type of pulsating variable star that exhibits a period-
luminosity law so that knowing the period of variation of the star allows
astronomers to determine its luminosity, thereby making it possible for
astronomers to calculate its distance from Earth.
Cerenkov radiation (pronounced “Chair-in-koff ”). Electromagnetic radia-
tion that is a result of high-energy, nearly massless particles, called neutrinos,
interacting with normal matter.
clusters of galaxies. Groups of more than 50 and up to 1,000 galaxies; clusters
are 2 to 3 Megaparsecs in diameter.
cold dark matter. Slow-moving matter that does not radiate any electromag-
netic radiation, but has mass.
constellation. A group of stars defined by the International Astronomical
Union that define a region in the sky.
contact binary. A binary system of stars where the two stars are touching one
another.
continuous spectrum. A type of spectrum that includes all wavelengths of
visible light. This looks like a rainbow.
convective transfer. The transfer of energy by convection or the movement
of material that contains thermal energy; the material is heated, making it
Glossary • 193
less dense and causing it to rise; as it rises, it radiates heat and becomes more
dense, sinking.
dark matter. Matter that does not radiate any electromagnetic radiation; as
yet, it is not known what type of matter this might be.
dark nebulae. Clouds of dust that block the light of stars and gas behind
them.
declination (Dec). The celestial sphere equivalent of lines of latitude on Earth
measured in degrees; the line of zero degrees is a projection of Earth’s equator
on the sky and is called the celestial equator.
differential rotation. A mode of rotation about a center where every part
moves at a different angular rate around the center, but with the same
linear rate.
diffraction. The spreading out of light due to the deformation of a wave as it
passes through an aperture.
disk. The flattened part of a galaxy that contains most of the gas and dust. A
flattened shape of gas and dust (as in accretion disk).
disk galaxies. Galaxies that have a flat disk-like component that dominates
the shape of the galaxy.
dissociate. To break a molecular bond, as in: molecular hydrogen dissociates
into atomic hydrogen due to high-energy photons.
Doppler boosting. A relativistic effect that causes ejecta moving away from
distant active galactic nuclei to appear to be moving at speeds greater than
the speed of light.
Doppler Effect. A phenomenon of waves that occurs when either the source
or the observer is moving; motion away results in longer wavelengths
observed than emitted, while motion towards results in shorter wave-
lengths observed than emitted.
dwarf spheroidal galaxies. Very small galaxies that appear to be roundish in
shape and contain little or no gas or dust.
dwarf star. A type of star that is small; can be either red dwarf (a type of main
sequence star), brown dwarf (a failed star), white dwarf (a type of hot stellar
remnant that does not fuse nuclei to produce energy), or a black dwarf (a
cooled off white dwarf that no longer emits any light).
ejecta. Materials that are ejected by an active galactic nucleus or a star.
electromagnetic radiation. All radiation that is in the form of photons; radia-
tion caused by the acceleration of charged particles, and is the interaction of
electric and magnetic fields.
electromagnetic spectrum. The range of wavelengths and/or energy of elec-
tromagnetic radiation.
elliptical galaxies. Spheroidal galaxies or galaxies that are roundish in shape
and contain little or no gas and dust.
emission spectrum. A type of spectrum that includes only a small fraction of
the wavelengths of visible light (those included are determined by the chemi-
cal composition of the emitted medium).
epicycles. Loops in an otherwise circular or elliptical orbit.
194 • GLOSSARY
event horizon. The point of no return around a black hole; beyond the event
horizon approaching the black hole, the escape velocity exceeds the speed
of light.
extinction. The process of attenuating light; light that is either absorbed or
scattered is considered extinguished.
flocculent galaxies. Galaxies that have no dominant spiral structure, but have
bright, fleecy spurs of star formation throughout their disks.
gamma ray. The most energetic, shortest wavelength form of electromagnetic
radiation.
giant elliptical galaxies. The largest of all galaxies, these appear to be round-
ish in shape and contain little or no gas and dust.
giant star. A type of star that is very large and fuses helium into carbon.
globular cluster. A dense, spheroidal cluster of 10,000 to 1,000,000 stars, usu-
ally found in the halo of a galaxy.
grand design galaxies. Galaxies that have strong spiral arm structure, usually
two bright arms that extend from the bulge to the outer limits of the disk.
gravitational lens. A massive object that bends the space-time continuum
enough to change the path of light.
ground state. The lowest energy an electron can have; the closest (to the
nucleus) orbital an electron can be in around its nucleus.
groups of galaxies. Groups of less than 50, but more than 3 galaxies; groups
are 1 to 2 Megaparsecs in diameter.
halo. The spheroidal component of a disk-dominated galaxy that extends
beyond the visible disk component and contains the globular clusters and is
dark matter dominated.
horizontal branch. The second stage of the red giant phase denoting stable
helium fusion in the core.
H-R Diagram or Hertzsprung-Russell Diagram. A graph of stars with tem-
perature or spectral type on the x-axis and luminosity or absolute magnitude
on the y-axis; this diagram can be used to infer the evolutionary sequence
of stars.
infrared. The region of the electromagnetic spectrum that includes wave-
lengths just longer than visible light.
instability strip. A region that runs through the red giant, horizontal and
asymptotic branches where stars are unstable and, due to their oscillations,
become variable stars, varying their brightnesses over periods of hours
to years.
interference. The interactions of waves with one another; this interaction
can be additive (constructive) or subtractive (destructive) causing light and
dark spots.
interstellar medium. The material between the stars within a galaxy—usually
gas (hot, warm, and cold) and dust.
intracluster medium. The material found between galaxies within a cluster.
ionize. To lose an electron, as in: atomic hydrogen is ionized by ultraviolet
radiation.
Glossary • 195
Keplerian motion. A mode of rotation about a center where the objects that
are closer move faster and objects that are farther move slower.
look-back time. The phenomenon that objects that are far away are seen as
they were long ago.
low surface brightness galaxies. Galaxies that have a lower density of stars in
their disks; as such, these galaxies are not very bright compared to the night
sky, so they are difficult to detect.
luminosity. A measure of the amount of energy per second given off by a
light-emitting source (usually the energy is emitted by a star or some other
astronomical object).
main sequence star. A type of star that fuses hydrogen into helium.
metallicity. A measure of the amount of elements heavier than helium in any
celestial object.
microwave radiation. The region of the electromagnetic spectrum between
infrared (the region that includes wavelengths just longer than visible light)
and radio (the least energetic and longest wavelength electromagnetic
radiation).
Milky Way. The name for the galaxy in which we live.
molecular clouds. Large conglomerations of molecular gas (usually hydro-
gen or carbon dioxide).
nebula. A cloud-like celestial object that may be gas heated by stars, dust
obscuring starlight, dust reflecting starlight, the expanding outer shells of a
dying star, or distant galaxy or group of stars; plural: nebulae.
neutrino. A nearly massless, structureless subatomic particle; a type of
lepton.
neutron star. A type of stellar remnant that gives off light; its composition is
thought to be neutrons since its density is so great that electrons and protons
would not be able to be separate particles; these stars do not fuse nuclei to
produce energy; the energy emitted by these stars is thermal energy.
open cluster. A loose association of stars, usually found in the plane of a
galaxy.
parallax. A way to measure distance to relatively nearby objects using the fact
that observations from opposite ends of a baseline result in different posi-
tions of nearby objects relative to more distant objects. The change in appar-
ent position of a nearby object relative to more distant objects; measured in
tenths or hundredths of an arcsecond for stars.
parsec (pc). A unit of distance equal to 3.6 x 108 m or 3.26 light years.
photoionization. The process of ionization (losing an electron) by way of
interaction with electromagnetic radiation (photons).
photon. A particle of light; photons are massless, but contain energy; the
amount of energy a photon has determines the classification of the electro-
magnetic radiation.
planetary nebula. A cloud-like formation that is caused by the outer layers of
intermediate mass star, like our Sun, being ejected at the end of its final stages
of the red giant phase of its evolution.
196 • GLOSSARY
solid-body rotation. A mode of rotation where every part has the same angu-
lar rate around the center, but different linear rates.
spectral type, (also spectral class). A classification of a star’s spectrum that
can be used to identify its surface temperature.
spheroidal galaxies. Galaxies that are roundish in shape and whose constitu-
ents exhibit tri-axial motion (motion in all three spatial dimensions).
spin-flip transition. An energy transition that involves the flipping of a
subatomic particle that has the quantum property of spin (a proton or an
electron).
spiral density waves. The phenomenon of over-dense areas of interstellar
material that causes spiral arm patterns in some disk galaxies.
spiral galaxies. Galaxies that are disk-dominated and usually exhibit some
spiral structure within the disk, commonly known as spiral arms; the Milky
Way is a spiral galaxy.
star-forming regions. Areas within a galaxy where very young stars are pre-
sent; usually identified by ionized hydrogen gas.
starbursting galaxies. Galaxies that are undergoing a burst of star formation—
usually indicated by an excess of ionized hydrogen emissions.
Stefan-Boltzmann Law. A mathematical relationship describing the depen-
dence of luminosity on temperature and radius of the source.
supercluster complexes. The largest structures in the universe made of
superclusters.
superclusters. Large structures made of clusters of galaxies.
supergiant. A type of star that is extremely large; this type of star fuses ele-
ments heavier than hydrogen into elements heavier than helium in its core to
maintain equilibrium.
superluminal speed. A speed greater than the speed of light (2.99 x
105 km/s).
supermassive black hole. A singularity usually located at the center of a galaxy
that contains a mass greater than about 1 million times the mass of the Sun.
supernova. An explosive event that marks the end of a massive star’s exis-
tence; this occurs when thermonuclear fusion can no longer occur in the
core of the star so that its outer layers collapse on its core.
synchrotron radiation. Electromagnetic radiation emitted by charged par-
ticles traveling at relativistic speeds (close to the speed of light).
T Tauri. A type of variable star that is associated with Sun-like stars in their
pre-main sequence stage of evolution.
terrestrial. Anything that is found on Earth or due to Earth’s atmosphere is
characterized as terrestrial (as opposed to solar, or interstellar, for example).
thermonuclear fusion. The fusion of light-weight elements to form heavier
elements occurring under conditions of extremely high temperatures and
extremely high densities.
torus. A doughnut shape; usually this shape is associated with the clouds of
dust and gas surrounding an accretion disk in the central engine of an active
galactic nucleus.
198 • GLOSSARY
triaxial. Something that uses all three spatial dimensions; as in, triaxial
motion.
triple alpha process. A process for fusing helium into carbon using three
steps and three alpha particles (a.k.a. helium nuclei).
turn-off point. The point on a color-magnitude diagram of a cluster where
the main sequence turns off to the red giant branch.
ultra-compact dwarf galaxies. Extremely small dense groups of stars, gas,
and dust considered to be some of the smallest and faintest galaxies in the
universe.
ultraviolet. The region of the electromagnetic spectrum that includes wave-
lengths just shorter than visible light.
x-ray. The region of the electromagnetic spectrum between gamma rays (the
most energetic and shortest wavelength) and ultraviolet (the region that
includes wavelengths just shorter than visible light).
Bibliography
Aguilar, David. Planets, Stars, and Galaxies: A Visual Encyclopedia of Our Universe.
Washington, DC: National Geographic Children’s Books, 2007. An excellent
volume for middle school and high school students.
Asimov, Isaac. Asimov on Astronomy. New York: Bonanza Books, 1988.
Bennett, Jeffrey D., Megan Donahue, Nicholas Schneider, and Mark Voit. The Cos-
mic Perspective. 5th ed. San Francisco: Benjamin Cummings, 2007.
Binzel, Richard P., series ed. University of Arizona Space Science Series. Tucson: Uni-
versity of Arizona Press, 1979.
The University of Arizona Press’s long-running Space Science Series publishes
cutting-edge research on planetary sciences. This series is relatively technical,
aimed at graduate school–level students and designed as a general reference for
professionals. The series is currently 30 volumes in total, but the books in the
series relevant to the topics discussed in this volume are listed below. The series
is periodically updated, so, for example, Protostars and Planets V supersedes
Protostars and Planets IV; however, earlier versions of books can still provide
interesting and useful information and so are also listed below.
Planets, Stars and Nebulae Studied with Photopolarimetry. Gehrels, T. 1974.
Protostars and Planets III. Levy, Eugene, and Jonathan I. Lunine, with the assistance
of Mildred Shapley Matthews, and Mary L. Guerrieri. 1993.
Protostars and Planets IV. Mannings, Vince, A. P. Boss, and S. S. Russell. 2000.
Protostars and Planets V. Reipurth, Bo, David Jewitt, and Klaus Keil. 2007.
Chaisson, Eric, and Steve McMillan. Astronomy Today: Stars and Galaxies. Vol. II.
6th ed. Menlo Park, CA: Benjamin Cummings, 2007.
DeGrasse Tyson, Neil, Charles Tsun-Chu Liu, and Robert Irion. One Universe: At
Home in the Cosmos. Washington, DC: Joseph Henry Press, 1999.
Elmegreen, Debra Meloy. Galaxies and Galactic Structure. Englewood Cliffs, NJ:
Prentice Hall, 1997.
Freedman, Roger, and William J. Kaufmann III. Universe. 8th ed. New York: W.H.
Freeman Company, 2008.
Hawking, Stephen. On the Shoulders of Giants. Philadelphia: Running Press, 2002.
Jastrow, Robert. Red Giants and White Dwarfs. New York: W.W. Norton and
Company, 1990.
Ratay, Douglas L., Multi-wavelength Observations of Barred, Flocculent Galaxies.
Available at: http://etd.fcla.edu/UF/UFE0005401/ratay_d.pdf.
199
200 • BIBLIOGRAPHY
Rubin, Vera C. Bright Galaxies, Dark Matters. New York: AIP Press, 1996.
Seeds, Michael A. Astronomy: The Solar System and Beyond. 5th ed. Pacific Grove,
CA: Brooks Cole, 2006.
———. Stars and Galaxies. 6th ed. Pacific Grove, CA: Brooks Cole, 2007.
WEB SITES
Absolute magnitude, 39, 40, 42, 45, Atomic hydrogen gas, 79, 82, 108, 136
51, 57, 60, 61, 87, 116 Autumnal Equinox, 2, 4
Absolute zero, 7
Absorption spectrum (lines), 27, 29, Balmer series, 152
30, 31, 35, 82, 90, 92, 101, 171 Bands, 51
Accretion disk, 63, 79, 80, 91, 92, 93, Bars, 127, 141, 166, 167, 168, 188;
94, 95, 109, 138, 150, 168, 171, barred spiral galaxy, 119, 122–23,
172, 173, 177, 178, 184 166–68; central bar, 122, 145, 147,
Active galactic nucleus (AGN), 148; peanut-shape bar, 122, 167;
168, 171, 175, 176, 177, 178, 184, weak bar, 16, 122
188 Barycenter, 97
Active galaxies, 111, 171–79 Bayer designation, 8
Alcor, 99 Bellatrix, 8
Aldebaran, 8 Beta decay, 65–66, 158, 159
Algol, 95 Beta Lyrae, 95
Altair, 8 Betelguese, 8
Andromeda, 132, 165–66 Big Dipper, 3, 38, 41, 99
Anthropocentric, 97–98 Binary star system, 46, 81, 92, 96, 97,
Apparent magnitude, 39–40, 41, 43, 98–103
45, 51, 57, 58, 60, 87, 116 Blackbody, blackbody spectrum,
Aquila, 3 blackbody curve, 15–20, 21, 23,
Arcsecond, 44 27, 112, 113, 138, 158, 173
Arms (spiral), 120, 121, 122, 136–40, Black dwarfs, 74, 157
145–60 Black hole, 76, 77, 78, 79, 88, 99, 105,
Arp, Halton, 127 137, 138, 141
Asterism, 3 BL Lac objects (Blazars), 176, 179,
Astrologers, astrology, 4 184; BL Lacertae, 176
Astrometric binary, 99, 101, 102 Blueshift, 184
Astrometry, 99 Bohr, Christian, 29
Astronomical unit, 44 Bohr, Ellen, 29
Asymptotic giant branch, 59, 61, 62, Bohr, Neils, 29
72, 73, 89, 92 Bohr Model, 27, 28, 108, 109, 151
201
202 • INDEX
Lagrange, Joseph-Louis, 105 Nebula, 63, 77, 117, 124, 130, 132,
Large Magellanic Cloud, 127 135, 139
Late type spirals, 122 Neutral atomic hydrogen, atomic gas
Lenticular galaxies, 126–27 cloud, 138, 150, 151, 152
Light-year, 184 Neutrino, 66, 158; neutrino
Light curves, 125 detectors, 159–60
Light pollution, 38 Neutron degeneracy, 76
Index • 205
Neutron stars, 8, 76, 77, 78, 83, 105, Protoplanetary disk, 91, 150; proto-
138 planetary nebulae, proplyds, 80, 91
Nova, 80–81, 85, 90, 93, 94 Protostar, 80, 83, 137
Nuclear ring, 167 Proxima Centauri, 125
Pulsar, 78
Occam, William of, 120 Pulsating variable star, 62, 85, 86–90
Occam’s Razor, 119, 120
Omicron Ceti, 89 Quasar, 175–76, 177, 179, 184
Open clusters, 56, 57, 8, 103, 135–36,
137, 142 radial motion, 165
Orion, 3, 8, 137, 139; Orion Nebula, radiative transfer, 68, 109, 111, 112;
131, 137, 139 radiative zones, 105
Other galaxies, 118, 126–27, 186 radio, 14, 78, 110, 111, 139, 151, 171,
Outer ring, 166, 167 173, 175, 176, 177
radio galaxies, 173–75, 179, 184
Pairs of galaxies, 163–65 radio lobes, 173, 174, 177, 178
Parallax, 25, 44, 57, 117, 125 radius of corotation, 166–67
Parsec, 40, 44, 117 R Cassiopeiae, 8
Pattern speed, 167 R Coronae Borealis variable star,
Pauli Exclusion Principle, 70 91–92
Payne-Gaposchkin, Cecilia, 34, 48 red dwarf stars, 91, 93, 165
Peculiar galaxies, 127 red giant branch, 59, 61, 62
Period-luminosity law, 57, 132 red giant stars, 52, 61, 75, 86, 87, 94,
Photographic film, 14 141, 165
Photoionization, 138 redshift, 183, 184
Photometer, 51 reflection, 14
Photon, 15, 28 reflection nebula, 139–40
Physical binary, 99, 101 refraction, 14, 25
Pickering, Edward Charles, 31, 47 resonance, 167
Planck constant, 15 Rigel, 8
Planetary nebula, 73, 79, 83, 138 Ring Nebula, 73
Planets, 3 Roche Lobe, 81
Pleiades, Seven Sisters, 57, 60, 61, 139 Römer, Ole, 25
Polaris, 8 rotating variable star, 85, 95–97
Pollux, 8 rotation curves, 150–60
Polycyclic aromatic hydrocarbons RR Lyrae, 57, 85, 87
(PAHs), 111, 112–13 Russell, Henry Norris, 34, 47, 48
Population I stars, 88 Rutherford, Ernest, 29
Population II stars, 87
Pre-main sequence evolution, 64; Saha, Meghnad, 33–34, 48
pre-main sequence stars, 63–65, Satellite galaxies, 123
93 Seyfert, Carl, 171
Progenitor, 81, 83 Seyfert galaxy, 171, 172, 175, 184
Proper motion, 3 Seyfert Type 1, 172, 178
Proto-galaxies, 186, 187, 188 Seyfert Type 2, 172–73, 178
206 • INDEX
Lauren V. Jones completed her A.B. in both physics and astronomy at Vas-
sar College, where she worked with Cindy Schwarz and Deborah Elmegreen.
She earned her first M.S. in physics, with specialization in astronomy, at
Moscow State University, where she worked with Anatoly Vladimirovich
Zasov on the source of far infrared radiation in galaxies. Jones then returned
to the United States to work with William Keel at the University of Alabama
on her second M.S. in physics with specialization in astronomy. Her thesis
incorporated more recent data and deconvolution techniques to analyze the
contributions of different stellar populations to the far infrared emissions
observed. She next completed an M.S. in astronomy and a Ph.D., working
with Richard Elston, at the University of Florida. For five years after com-
pleting her Ph.D., Dr. Jones taught at different institutions of higher educa-
tion. Becoming interested in science education research while working with
Timothy Slater and Edward Prather at the University of Arizona, she took a
position with the Ohio Department of Education.